CALIFORNIA TUMOR TISSUE REGISTRY

NINETY-FIRST SEMI-ANNUAL. SLIDE SEMINAR

ON

BREAST TUMORS

MODEjt4TOR:

DAVID L. PAGE, M. D. PROFESSOR OF PATHOLOGY

DIRECTOR OF ANATOMIC PATHOLOGY VANDERBILT UNIVERSITY NASHVILLE, TENNESSEE

CHAIRMAN:

RICHARD L. JOHNSON, M. D. ASSOCIATE PATHOLOGIST

HUNTINGTON MEMORIAL HOSPITAL PASADENA, CALIFORNIA

SUNDAY - DECEMBER 8, 1991 9:00A.M. - 5:00 P.M.

REG! STRATI ON: 7·: 30 -A .M.

SHERATON EAST HOTEL ON HARBOR ISLAND SAN DIEGO, CALIFORNIA

(619) 291-2900

Pl ease bring your protocol, but do not bring sl ides or microscopes to the meeting.

CONTRIBUTOR: Burr Hartman, 0. 0. OECE~tBER 1991 - CASE NO. 1 Lancaster, Cali fornia

TISSUE FROM: Breast ACCESSION NO. 26351

CLINICAL ABSTRACT :

His tory: This 48-year-o l d white fema l e with kno~m severe emphysema was admitted t o the hopsital on September 20, 1988 with history of right fluctuan t breast mass for the past three years. The mass was needled 2-3 times with negative results during hospitalization a month ago for respiratory arrest.

However, on return to the Outpatient Department for evaluation, the mass was about 6 em. in diameter and hard. Axillary nodes were present. Mammogram showed mi crocal cifications and general .configuration compatible wi t h carcinoma until proven otherwise.

Physical examination: The right breast appeared with a mass in the lower ou ter quadrant, 6-8 em. in diameter, hard, poorly mobi le, and fixed to the deeper layers but not producing significant skin changes . There was no nippl e discharge. One 1 ymph node, 2. 5 em. in diameter, was pa 1 pated in the axi 11 a.

SURGERY : {September 20, 1988)

Under l ocal anes thesia, an en bloc resection of the mass was attempted. It was somewhat bloody because the tumor was found to be very vascularized. When all of the tissue was thought to be removed, some tissue was found stuck to the thoracic wall. Thi s was removed l eavi ng what appeared to be normal tissue.

GROSS PATHOLOGY:

The specimen consisted of three f ragments of yellow-tan fibrofatty breast t issue, measuring 4.5 x 4.5 x 3.0 em., 5 x 2.5 x 2.0 em., and 4.5 x 4 x 2.0 em. Multisectioned surfaces revealed a red, firm, fleshy process with indistinct ~~rgins . In one of t he biopsy fragments, there was a red-fleshy, grossly well circumscribed nodule, measuring 0.8 em. in diameter.

CONTRIBUTOR: William Worthen, M. D. DECEMBER 1991 - CASE NO . 2 Inglewood, California

TISSUE FROM: Breast ACCESSION NO. 26641

CLINICAL ABSTRACT:

Historl : This 86-year-old Caucasian fema le was admi tted to the hospital on May 23,989 by way of the Emergency Room because of a severe fecal impaction. Thi s apparently started some 4-6 weeks ago with constipation, nausea and vomiting with increasing abdominal di stens ion, intermittent abdominal pai n with loss of appetite . An enlarged right breast was noted.

Past hi story: Relevant to the breast problem, the patient stated t hat she had a breast mass since age 18 y~ars and paid no significant attention to it and refused any kind of surgical procedure. It had remained the same size until recently when it increased in size .

Physical examination : The patient was extremely dehydrated and mal ­nourished . The abdomen was massi vely distended with l arge rolling masses through· out the abdomen . A rectal examination revealed a huge fecal impaction.

After therapy for correction of the breast: "Huge 1 umpy mass f i 11 i ng entire area whi ch threatened possi bl e rupture. in greatest diameter. "

SURGERY: (June 5, 1989)

A si mpl e mastectomy was performed .

GROSS PATHOLOGY:

impaction, attention was paid to the breast. There appeared to be a f luctuant It measured approximately 7 x 8 em.

The breast weighed 250 gm. with a skin ellipse, 14 x 12 em. The nipple was flattened and centrally placed. Cut section revealed a solitary, ovoid, sharply demarcated gray-white mass, 6.0 x 5.5 x 5.0 em. The majori ty of the mass had a soft, fl eshy, gray-white cut surface . Centrally located was a hemorrhagic "cysts " with smooth walls, 4. 5 x 2.5 x 2.5 em., filled with clotted blood.

CON TRIBUTOR: Richard L. Johnson, N. D. DECEMBER 1991 - CASE NO. 3 & 4 Pasadena, California

TISSUE FROM: Left breast ACCESSION NO . 27010

CLIN ICAL ABSTRACT:

History: This 15-year-old Black girl was seen in July 1991 complaining of enlargement of her left breast for about two months . She stated it has always been slightly larger t han the r ight, but over the last two months noted dramatic i ncrease, as wel l as t he nippl e . She denied any pain in the breast.

Physical examination: The left breast was remarkable for massive hyper­trophy. It was approximately three times the size of the right breast. A very large mass was palpated which was not fixed to the skin or chest 1~all. The edges were di ffi cul t to define· clearly, and whether there ~1as lobulation or actual ly several masses close to each other . The nipple-areolar complex was also massively enlarged and at the 2:00 o'clock position there was a 5 mm. single firm, discrete subcutaneous nodule. The right breast was of moderate size and had a 5 mm. smooth firm nodule relatively high on the breast at the 12:00 o'clock posi t ion.

Mammogram was interpreted as normal, however ultrasound showed a 13 em. smooth-edged solid mass, occupying the majority of the left breast.

SURGE RY: (August 22, 1991}

Partial mastectomy with reconstruction of left breast and excisional biopsy of the right breast were performed.

GROSS PATHOLOGY:

The left breast mass was a 418 gm., firm, bosselated mass, measuring 13 x 10.5 x 5 em. The cut surface was compartmentalized into confluent nodules, ranging from 3-7 em. in di ameter. One nodule did contain zones of a frond-like configuration . The cut sur faces showed a fi rm, pink-t an appearance wit hout evidence of necr osis.

CONTRIBUTOR: Nelson Qui gley, M. D. DECEMBER 1991 - CASE NO . 5 Anahiem, California

TISSUE FR0~1 : Right breast ACCESSION NO. 26239

CLINICAL ABSTRACT:

History: This 57-year-old Caucasian femal e was f i rst seen by the admitting physician on February 22, 1988 with history of painless right breast mass, 2 months' duration. She was admitted to the hospital on t~ay 5, 1988 following a xero mammogram (February 16, 1988) and chemotherapeutic three cycle treat­ment.

Physical examination: Hell developed, ~1ell nourished obese, jovial patient who appeared to be in good general co~dition . The body systems other than for the breast were without significant findings . An 11 em. hard centra 1 mass with ni ppl e retraction was present in the right breast wi th erythema and edema over t he overlying skin. Following chemotherapy, there was marked reduction in edema surrounding the nipple and areola as well as disappearance of the erythema . The axilla was without lymphadenopathy . The left breast and axil la were norma 1 .

SURGERY : (May 9, 1988)

A r ight modified radical mast ectomy was performed . The most superi or axillary lymph node appeared to be i nvol ved.

GROSS PATHOLOGY:

The breast measured 35 x 16 em., covered by an el lipse of skin, 25 x 10 em. with an inverted nipple . In the para-areolar region, there was an ovoid, gray, ill-defined mass, 7 em. in greatest diameter. Multisectioned tissue was gray-white , with an ill -defined tumor that was firm. Axillary fat pad revealed "several" lymph nodes, some of which were grossly involved with t umor. The highest separatel y submitted lymph node, 1.5 x 1 x 0.7 em., ~1hen sectioned had a yellow-gray color.

CONTRIBUTOR : James Tenney, M. D. DECEMBER 1991 - CASE NO. 6 Bishop, California

TISSUE FROM: Breast ACCESSION NO. 27013

CLINICAL ABSTRACT:

History : This 41-year-old female on routine physical examination 1~as asymptomatic except for mild fibrocystic disease and a routine mammogram was performed.

Physical examination was unremarkable . The breast and axilla revealed no palpable lesions .

~1ammography: t4i croca 1 ci fi cation~ in the right breast were sus pi ci ous for care i noma .

SURGERY: (August 29, 1991)

Biopsy with mammographic local ization of the calcific identified area was performed.

GROSS PATHOLOGY:

Breast t issue, 7 x 4 x 4 em. , was received with a metallic wire in place. The tip of the wire t issue calcification was noted grossly. Multisectioned t issue was homogeneous wi th a few small cysts.

CONTRIBUTOR: Mindy Cooper-Smith, M. D. DECEMBER 1991 - CASE NO. 7 Ojai , California

TISSUE FROM: Breast ACCESSION NO. 25793

CLINICAL ABSTRACT:

History: This 59-year-old Caucasian femal e who developed an indurated area in her right breast 1t years ago with a history of having a mammogram while in "Idaho". Following a breast examination , the physician advised t he patient to be fo llowed. Subsequently, she was admitted to the present hospital in California on August 8, 1986 where a right breast mass was confirmed on physical examination and by mammography. Ultrasonography of the palpable nodule showed internal echoes i ndicating a solid rather than cystic lesion . There was some through transmission which suggest ed a fibroade noma. However because of its solid nature, a biopsy was performed on August 25, 1986. Her final admiss ion was on September 16, 1986 .

Physical examination was essentially unremarkable except for the right breast. There was an indurated mass underlying a healed biopsy scar in the right upper quadrant . The axi l la revea led no pal pable nodes.

SURGERY : (September 17, 1986}

A simple mastectomy was performed .

GROSS PATHOLOGY:

The right breast measured 18 x 15 x 3 em. with a 15 x 10 em. ellipse of overlying skin. An incompletely healed surgical scar, 3 em. from a grossly unremarkable nipple and areola 1·1as present. This was opened into a cavity of fat necrosis with hemorrhage situated 1 em. from the deep margin of resection . When mul tisectioned, there were no areas cons istent with the previously biopsied t i ssue. Mul t iple 2-5 mm. in diameter thin-wal led fluid cysts were present. The posterior margin of the resection was int act and no skeletal muscle was identified .

CONTRIBUTOR: William Snider, M. D. DECEMBER 1991 - CASE NO. 8 West Covina, Cal ifornia

TISSUE FROM: Breast ACCESSION NO . 16690

CLINICAL ABSTRACT:

History: This 68-year-old diabetic mal e noted enlargement of the right breast for approximately one year ago. A xanthochromic nipple discharge began 2-3 months ago.

Physical examination: Other than a moderate degree of HHD and ASHO, the patient was physically in good condition.

Breast: There was a movable ~ass approximately 7 em . in greatest dimension .

SURGERY: (November 1967)

A si mple subcutaneous surgical removal of the nodular lesi on was done .

GROSS PATHOLOGY :

The specimen was an 8. 5 x 7 x 4 em. mass of rubbery, pale tissue inter­spersed with yell ow fatty tissue which 1~hen sectioned revealed many circum­scribed bulging blue-gray nodular areas. These ranged in size up to 1.0 em . in diameter .

CONTRIBUTOR: Bahram Parsa, M. 0. DECEMBER 1991 - CASE NO. 9 Los Angeles, California

TISSUE FROM: Breast ACCESSION NO . 2700B

CLINICAL ABSTRACT:

History: The patient was a 22-year-ol d black fema l e 1~ho had lumps in the right breast for approximately one year. She stat ed that she had masses in similar areas during her pregnancy 3 or 4 years ago which subsequently resolved and did not return again until this past year. She complained of local discomfort with compression .

Physical examination revealed two palpable well defined masses in the right breast, measuring 2 em. in diameter each at the 10 o'clock and 12 o'ocTock positions. They were mobile, firm, and non-tender . There was no axil lary adenopathy on the right. The left breast was normal.

SURGERY: (July 10, 1991 )

Excisional biopsies of the masses were performed. The masses appeared to be fibrocystic disease and there was no clear cut evidence of fibroadenoma .

GROSS PATHOLOGY :

The specimen consisted of two pieces of lobulated yellow to grayish tissue, equal in size, each measuring 4 x 3.5 x 3 em., and was f i rm to rubbery in con­sjstency. Cut section showed granular grayi sh dirty-looking cut surface with microcystic structure.

CONTRIBUTOR: Del R. Dickson, M. D. DECEMBER 1991 - CASE NO. 10 Santa Barbara, California

TISSUE FROM : Breast ACCESSION NO. 27046

CLINICAL ABSTRACT:

History: The patient with known hypochromic anemia since 1978 was seen in the doctor's office for a yearly check up. Routine mammogram was done whi ch showed calcification, suspicious for malignant .

SURGERY: (June 26, 1991)

An excisional biopsy was performed.

GROSS PATHOLOGY:

The specimen consisted of a fatty portion of breast tissue, 36 x 30 x 10 mm. It had a wi re protruding from the specimen. No mass was i dentified on cut section.

CONTRIBUTOR : E. DuBose Dent Jr. , M. D. DEW·1BER 1991 - CASE NO . 11 Glendale, California

TISSUE FROM: Tail of breas t ACCESSION NO. 26938

CLINICAL ABSTRACT:

History: This 48-year-old female was fi rst seen i n a physician's office on November 5, 1990 1~i th the complai nt of a l ump in the right breast of 3 years ' duration. This had become tender and larger in the last three months.

A mammogram done a few days previously at the Cancer Center reported as 10 em. dominant mass, right upper quter quadrant , representing ei ther a cyst or a 9iant f ibroadenoma . Further evaluation by ul trasound revealed a l arge discrete well marginated solid mass, which was compressible and varied i n shape, and fibroadenoma was unlikel y. On re-review of the mammogram, some of the better penetra ted views suggested the possibility of some central fa t wi thi n the tumor and rai sed the possibi lity of a so-called fibrolipoadenoma .

Phvsical exami nat ion: The breast reveal ed a large 10 em. cystic mass in the right upper quadrant. The remaining breast was multinodular. It ~1as aspirated and gelatinous flui d was obtained. This was submi tted for cytology and the patient was requested to return on January 22, 1991.

OFFICE SURGERY: (January 22, 1991)

Under local anesthesia, a soft, movable t issue mass was encountered on t he right upper external quadrant of the breast. Upon di ssection of t hi s mass, there was some colloid gelatinous type material inside of the mass that felt to be cystic. After the mass was exci sed, there was infiltration of the muscle by the same type of a mass. No dominant mass coul d be fou nd under the ax ill a.

GROSS PATHOLOGY:

The specimen was an irregular encapsulated mass, 10.0 x 9.0 x 7.0 em. Vihich v1hen sectioned was mucoid. The color varied from red-brown to yellow­bro~m t o gray-white.

CONTRIBUTOR: Dennis Kasimian, M. D. DECEMBER 1991 - CASE NO. 12 Van Nuys , California

TISSUE FROM: Breast ACCESSION NO. 26946

CLI NICAL ABS TRACT :

Hi story: Thi s 47-year-ol d black fema l e noted 5 months ago a lef t - si ded breast mass whi le t aking a shower . She did not seek medical hel p and the l esion seemed to grow over the next 5 mont hs and became painful to t ouch .

Past hi story: Gravida I , para I, Ab 0 and nursed her one child . She had a t otal abdominal hysterectomy for "cervi cal cancer". Ovaries were l eft intact. She i s premenopausal and on no hormones. She is being treated for symptoms of paranoid schi zophreni~ .

Physical examination was essentially without signifi cant findings except for a hard, wel l ci r cumscribed 5 em. mass in the l eft upper quadrant of t he breast. Some bilateral shotty lymphadenopathy was present without gross evidence of clinically significant nodes in the axilla . The cont ralateral breast was unremar kable .

Mammogram was performed 1~hi ch l i kewi se appeared t o be very suspicious in radi ographic appearance with a large, domi nant mass noted and numerous suspicious microcalcifications present within the upper outer quadrant of the left breast . An aspiration was performed which was highly suspicious for malig­nant cell s .

SURGERY: (January 15, 1991}

A left modified radical mastectomy was performed following the rapid frozen report.

GROSS PATHOLOGY :

The left breast weighed 740 gm. and measured 26 x 17 x 5 em. with attached 8 x 7 x 2.5 em . axi l lary contents. Attached was a 19 x 10 em. el l i pse of skin with an evert ed hyperpigmented 1. 5 em. ni pple and a 4 em. areola . There was an obl ique sutured 3. 5 em. incisi on midway al ong t he exc ised skin. Beneath the inci sion was a 4. 5 em. blood-soaked gauze . The shaggy margins of the cavi ty revealed in the depth a firm, near spherical mass, 5. 5 em. in greatest dimensions . The borders of t he tumor were fa i rly well del ineated from the surrounding hemorrhagi breast and extended within 0.1 to 0.2 em. from t he depth of the margin . It lay several centimeters bel ow t he ski n surface . On mul ti section, t he tumor revealed firm to friabl e t an to pink- tan t i ssue with extensive hemorrhage and necrotic softening . Within the axillary contents, there were multiple firm, ovoid , red-tan and gray-tan lymph nodes t hat measured up t o 1.5 em . in grea test dimension.

CONTRIBUTOR: Dennis Kasimian, M .D. DEC EMBER 1991 - CASE NO. 13 Van Nuys, California

TISSUE FROM: Breast ACCESSION NO. 24952

CLINICAL ABSTRACT:

History: This 82-year-old Mexican-American femal e underwent exc1s1on of an intraductal papil l oma of the left breast in 1979. Subsequently she developed a mass in t he area which was aspirated and cytol ogy reported to be negative . There was progressive increase in the size of the breast in this area with rapid growth in the last six months of the nontender hard mass. She was reluctant to have surgery and postponed it for several months but, after much consultation, agreed to proceed with surgery.

Physical examination: The left breast was extremely pendulous with an ol d scar over the medial upper quadrant at approximately 10 o'clock position. Deep to this scar was a somewhat movable, very large, relatively painless mass, 6 x 8 em. in diameter. This had been the site of obvious enlargement with stretching of the skin but no adherence to the underlying skin. There was one lymph node palpable in the left axilla. There was no nipple discharge.

SURGERY: (May 9, 1g83)

Under general endotracheal anesthes ia, a solid, encapsulated mass, 6 to 8 em. in its diameter was excised and a total left mastectomy without axillary dissection was carried out.

GROSS PATHOLOGY:

The tumor measured 8.5 x 7.5 x 6 em. and had a bosselated surface. The cut surface was firm, faintly nodular, and pink- tan to yellow. Some areas appeared consistent with cartilage or bone .

CONTRIBUTOR: Arthur L. Koehler, M. 0. h .s.a.oi<0.\\0., \.o.\\~(l'('\\\0.

TI SSUE FRO~!: Breast

CLINICAL ABSTRACT:

DECEMBER 1991 - CASE NO. 14

ACCESSION NO . 27014

History: She was seen on June 29 , 1989 with a history of a 2 x 2 em. left breast mass of several months' duration.

SURGERY: (August 3, 1989)

Under general anesthesia, the breast mass at 6 o'clock position was removed along with what appeared to be very severe cysti c disease with several hundred cysts within a few c~bic em.

GROSS PATHOLOGY:

The specimen consisted of a portion of irregular breast tissue, approximating 5 x 3 x 4 em. It was moderately firm due to the presence of multiple confluent cysts from 0.2 to 0.3 em. i n diameter whi ch contained variable amber to clear fluid . The adjacent stroma was quite soft in consistency with only a 3 em. area of linear fibrosis in which edematous cystic s tructures occurred. No linear striations were seen.

CONTRIBUTOR: Ph i lip Van Hale, M. D. DECEMBER 1991 - CASE NO. 15 Pasadena, California

TISSUE FROM: Breast ACCESSION NO. 27027

CLINICAL ABSTRACT:

History: This 58-year-old white female had noticed a left breast mass about two months ago. A mammogram revealed a 3 em. round mass in the left breast. At an office visit, a hard mass v1as present in the left upper quadrant. A needle aspi ra te was non-diagnostic. She was admitted to the hospi tal for an excisional biopsy on September 11, 1991. ·

Past history: She was a di abetic cystectomy and panniculectomy in 198~. contri butory.

witn hypertension. She had a chole­Familial history ~1as essentially non-

Phys i ca 1 examination revea 1 ed a we 11-deve 1 oped, we 11-nouri shed obese, pleasant woman in no apparent acute distress. The rest of the phys·ical was essentially normal including a blood pressure of 140/74. The ri .ght breast 1~as normal . The left breast revealed a. 3-4 em. hard mobile mass in the left upper quadrant· wi thout skin dimpling. The axi l la had no palpable nodes.

SURGERY: (September 11, 1991)

At the time of surge.ry the mass appeared to be moderately encapsulated and was excised in its entirety, although at one point it was entered and there appeared to be some necrotic material i n its center.

GROSS PATHOLOGY:

An ovoid segment of fibrofatty t i ssue that measured 6 x 6 x 3.5 em. in greatest dimension was multisecti oned. The central portion revealed an irregular, circumscribed polypoid mass that extended to. one margin. It was pi nk-white and measured 3.5 x 2.5 x 2 em. in greatest dimensions.

CONTRIBUTOR: James M. Tenney, M. D. DECEMBER 1991 - CASE NO. 16 Bishop, California

TISSUE FROH: Breast ACCESSION NO. 27029

CLINICAL ABSTRACT:

History: This 65-year-old white female was admi tted on Sept ember 30, 1991 for biopsy of a nipple lesion because of suspected Bowen's disease biopsied elsewhere. She noticed a year ago some i rritation ~1hich cleared up when she was given anti biotics for an upper respi ratory infection. It recurred in the form of an eczema ras h. Fami li al and past hi stor ies were unremar kabl e.

Physi cal exami nation was unreveal ing except for t he breasts 1~h ich were full , symmetrical with the l ef t n_i ppl e and areol a grossly infl amed when compared to the r ight breast. Palpation of the breast s revealed mi ld nodularity wi th­out any dominant masses . The axillary areas revealed no lymphadenopathy .

SURGERY: (October 1, 1991)

A removal of the nipple and areol a and t issue t oward the t ail of breast was performed.

GROSS PATHOLOGY :

A skin el l ipse with an erect nipple i n the central portion measured 9. 7 x 5.1 em. The ni pple meas ured 1. 3 em. in diameter and t he areola measured 4.3 em . in diameter . The nippl e was f ocally denuded, variegated tan-gray and blended wi th a t otal ly roughened and denuded areol a. The under surfa ce of the skin ellipse was fatty tissue, 9 x 4.5 x 3.0 em. Multi ple sections reveal ed fatty tissue and a separate t i ssue designated axillary t ail ~1hen sectioned revealed a "fatty" nodule .

CONTRIBUTOR: Gabriel Haiby, 11 . D. DECEMBER 1991 - CASE NO. 17 & 18 Encino , California

TISSUE FROM: Breast ACCESSION NO. 20104

CLINICAL ABSTRACT:

Hi story : Thi s 51-year-ol d female presented with a left breast mass meas uring 7 x 4 x 2 em. of unknown duration.

SURGERY: (February 1973)

An excisional biopsy of the left breast was performed .

GROSS PATHOLOGY:

The cut surface was firm, lobul ar, and contained cyst- l ike spaces through­out. These cysts contained semi-sol id mucoid material.

CONTRIBUTOR: Wi l l iam Hartmann, M. D. DECEMBER 1991 - CASE NO . 19 Long Beach, Cal iforni a

TISSUE FROM: Breast ACCESSION NO. 27041

CLINICAL ABSTRACT:

Histor y: Thi s 44-year-old fema l e presented in late September with a sel f- di scovered l ump in her right breast of one month 's durat ion.

Physical examination revealed a golfball size mass in the right breast.

Mammogram, August 16, 1991: There was a dense nodul e corresponding to the area of her complaint and the upper inner quadrant which measured approximately 1.5 em.

Ultrasound of the dense nodule showed mixed echogenicity with some cystic and some solid components. It was well mar~inated. It may represent a fibroadenoma or an usual ly proteinaceous cyst. (The cyst was aspirated and interpreted as unusual ? purulent cyst, right upper inner quadrant and may represent a breast abscess, though its position 1~as unusua 1) . ~

SURGERY: (September 27, 1991)

An exeisional biopsy was performed .

GROSS PATHOLOGY:

The specimen consisted of a 3 x 2.5 x 2.5 em. portion of breast tissue containing a discrete, 2.2 em. pink, papillary, fibro lesion surrounded by a capsule that measured 0.4 em. in thickness.

CONTRIBUTOR: William Hartmann, M. 0. DECEMBER 1991 - CASE NO. 20 Long Beach, California

TISSUE FROM: Breast ACCESSION NO. 27040

CLIN ICAL ABSTRACT:

History: This 50-year-old femal e presented with a history of silicone injections into her breast 25 years ago.

On phys ical examination, the breasts were described as being firm. The skin was purple and discolored throughout both breasts.

SURGERY : (October 11, 1991)

Modified mastectomies were performed.

GROSS PATHOLOGY:

The "right breast tissue'' was a tan-red fibroadipose tissue weighing 454 gm. and measuring 17 .0x 11.5 x 5.3 em. in greatest dimension. In addition, there were multiple smal l tan-red tissues weighing 37 gm. in aggregate. Palpation of the large specimen revealed a soft cyst ic- li ke center with multiple firm nodules of varying sizes on the margins. Serial section of the larger tissue revealed a pale tan, softened central area surrounded by a tan, glistening, firm, homogeneous tissue.

The "left breast tissue" was a large tan-red f i broadipose tissue measuring 15.0 x 14.0 x 5.0 em. in greatest dimensions and ~~eighing 491 gm. Serial sectioning demonstrated a cut surface simi lar to the right breast.

NINETY-FIRST SEf~I-ANNUAL SEMINAR ON BREAST TUNORS

DIAGNOSIS

Each registrant is urged to submit diagnosis in order to document a study of the cases in this slide seminar. This will simultaneously provide data for the Continuing Medical Education, California Medical Association, and will provide t he prelector with a l ist of items to be discussed in the differential diagnosis .

To permi t representative tabulations for the seminar, please mail diagnos is by Friday, December 6, 1991, to: Cal iforni a Tumor Tissue Registry, c/o Hungtington Memorial Hospital, 100 West California Boulevard, Post Office Box 7013, Pasadena, California 91109-7013.

Case No. 1 26351

Case No. 2 26641

Case No. 3 27010

Case No. 4 27010

Case No. 5 26239

Case No. 6 27013

Case No. 7 25793

Case No. 8 16690

Case No. 9 27008

Case No. 10 27046

Case No . 11 26938

Case No. 12 26946

Case No. 13 24952

Case No. 14 27014

Case No. 15 27027

Case No. 16 27029

Case No. 17 20104

Case No. 18 20104

Case No. 19 27041

Case No . 20 27040

PROGRESS IS OUR 110ST INPORTANT PROBLEM

SEMI-ANNUAL SEMINAR ON BREAST TUMORS DECEMBER 8, 1991

SAN DIEGO, CALIFORNIA

CASE 1 -Ace . No. 26351

Follow-up: She was taken back to surgery on September 30, 1988 for excisional biopsy of a 3 em. mobil e axillary lymph node wh ich appeared to be formed by the fusi on of two or three nodes. She expired 1/24/90 of respiratory fa ilure.

CASE 2 - Ace. No. 26641

Follow up: The patient recovered uneventfully from the surgery, but progressively devel oped increasing symptoms of senile dementia. She entered a nursing home and died of unrelated cases 2! years after the surgery .

CASE 3 & 4 - Ace. No. 27010

Follow-up : Patient returned January 6, 1992 because of a tender smal l palpable in the area of the previously removed tumor . Repeat mammogram ordered.

CASE 5 - Ace. No. 26239

Follow-up: Chemotherapy was completed in October 1988. She is alive and well as of November 26, 1991 and presently maintained on Tamoxifen.

CASE 6 - Ace. No. 27013

Follow-up: Repeat mammogram revealed more ca lci fications. She is scheduled to have re-excision of calci fica tion.

CASE 7 - Ace . No. 25793

Follow-up: Patient was last seen 12/4/1989 at which there was no re­currence she is being followed yearly with mammogram of left breast for mild fibrocystic change.

CASE 8 - Ace. 16690

No fol low-up.

CASE 9 - Ace . No . 27008

No fo 11 ow-up.

CASE 10 - Ace. No. 27046

No fo 11 01~-up.

CASE 11 - Ace .· No. 26938

Follow-up: A right breast segmentectomy, upper outer quadrant, with incontinuity axillary node dissection was performed on March 14, 1991 at City of Hope. There was no residual tumor and lymph nodes (22) were negative. As of November 1g91 she is doing fi ne and receiving no therapy.

CASE 12 - Ace. No. 26946

Follow-up: She received 4 courses of chemotherapy and was free of tumor. However, one month ago she was found dead of unrelated cause.

CASE 13 - Ace. No. 24952

Follow-up: She was hospitalized 9 months later at which time she was found to have lung and bone metastases secondary to primary breast sarcoma, right cranial nerve palsy secondary to medial sphenoid bone metastases, and right lytic lesion of the femur secondary to bone metastases from primary breast sarcoma . No current follow-up available.

CASE 14 - Ace. No. 27014

Fol low-up: She was seen in a $tudent health cl inic in March 1991 at wh ich t ime an ultrasound revealed a 5 x 9 mm. hypoechoic densi ty at t he 6 o' cl ock position of the left breast which may be of primarily cystic les.ion though some internal echoes were seen. The margins were not smoo t h and sharp. Follow-up by ultrasound recommended in 6 months.

She returned to her physician and when last seen on November 4, 1991, there was 1.5 x 2.5 em. thickening present, probable recurrent juvenile papillomatosis . She is to be re-examined in one month.

CASE 15 - Ace. No. 27027

Patient had a modified radical mastectomy on October 16, 1991 which was reported as no evidence of residual carcinoma. She is presently undergoing chemotherapy.

CASE 16 - Ace. No. 2702g

Follow-up: She subsequently had a mastectomy six days later which showed extensive intraductal atypical hyperplasia and carcinoma with 26 lymph nodes uninvolved.

CASE 17 & 18 - Ace. No. 20104

Follow-up: Re-biopsy was performed in June 1973. She was l ast seen in June 1976 with no evidence of res idual disease.

Case 19 - Ace. No. 27041

She tolerated radiation therapy 10/23/gl - 12/18/91.

CASE 20 - Ace. No. 27040

Follow-up: She went int o the hospital to have tissue expander put in which was removed and replaced with a prosthesis.

CALIFORNIA TUMOR TISSUE REGISTRY

NINETY-FIRST SEMI-ANNUAL SLIDES SEMINAR ON

BREAST TUMORS

David L. Page, M.D. Professor of Pathology

Director of Anatomic Patholog

Vanderbilt University Medical Center Nashville, Tennessee 37232

Sunday, December 8, 1991

San Diego

CASE NO.

1

2

3 & 4

5

6

7

8

9

10

11

12

13

14

15

16

17 & 18

19

20

MODERATOR:

DAVID L. PAGE, M. D. BREAST TUMORS

SHERATON EAST HOTEL ON HARBOR ISLAND SAN DIEGO, CALIFORNIA

DECEMBER 8, 1991

ACCESSION NO. DIAGNOSIS

26351 Complex intraductal carcinoma with infiltrating low grade, no special type carcinoma and multifocal infil­trating lobular carcinoma, pure and pleomorphic.

26641 Adenomyoepithelioma or tubular adenoma with myoid prominence.

27010 Giant Fibroadenoma with focal benign phyll odes tumor.

26239 Pleomorphic infiltrating lobular carcinoma with apocrine cytology.

27013 Non-comedo ductal carcinoma in-situ, micropapillary type.

25793 Adenoid cystic carcinoma.

16690 Multiple duct papillomas with focal encysted, non-invasive papillary carci noma and adjacent in-situ ductal carcinoma.

27008 Radical scar (complex sclerosing lesion)

27046 Amyloid deposits

26938 Myxoid liposarcoma

26946 Metaplastic carcinoma , chondrosarcoma

24952 Osteosarcoma

27014 Florid hyperplasia with cysts

27027 "Giant cell" carcinoma or anaplastic carcinoma.

27029 Paget's disease of the nipple.

20104 ~persecretory hyperplasia, cysts and xanthomatous duct ectasia.

27041 High grade, largely encysted papillary carcinoma, microinvasive

27040 Si 1 i cone granuloma

PAGE

1

2

4

6

8

9

11

13

14

15

16

19

20

21

22

24

25

26

CASE 1

Diagnosis: Complex intraductal carcinoma with infi ltrating low grade, no special type c arcinoma and multifocal infiltrating lobular carcinoma, pure and pleomorphic.

This case is a reflection of the complexity in patterns and space of many breast carcinomas. The margins of the tumor were somewhat ill-defined, probably because the tumor is a mixture of so many separate elements, dominantly in situ disease.

Both intraductal carcinoma and infiltrating lobular carcinoma are discussed below, but it should be mentioned here how important it is to adequately sample a mammary c arcinoma for the heterogeneity displayed in this neoplasm to such a marked degree.

The prognosis here is probably very similar for both the low grade no specia l type (or ductal) carcinoma and the pleomorphic lobular carcinoma. Although the growth fraction is low in both, it may well be lower in the infiltrating lobular carcinoma. Note that invasive lobular ca.rcinoma is diagnosed by intrinsic pattern criteria , and does not require lobular carcinoma in situ .

The invasive disease is evident in this case and may be measured at over 2-4 mm. at each site. More subtle contenders for invasion are often over-diagnosed as invasive disease when there is irregular invol vement of lobular units . Two recent papers document that cases signed out as "microinvasive" in the early 1980's behave like in s itu only disease 1,2. ·

References

l . . wong JH,Kopald KH and Morton DL: The impact of microinvasion on ax1llary node met astases and survival in patients with intraductal breast cancer. Arch.Surg. 125:1298-1302, 1990.

2. Rosner D,Lane WW and Penetrante R: Ductal carcinoma in situ with microinvasion: A curable entity using surgery alone without need for adjuvant therapy . Cancer 67: 1498- 1503, 1991.

CASE 2

Diagnosis: Adenomyoepithelioma or tubular adenoma with myoid prominence.

There has been a recent increase in interest in the co- occurrence of myoepithelial differentiation within other lesions of - the breast . Actually there is also an increase in the number of reports of individual breast fells with mixed luminal and myoepithelial differentiation , and myoepithflial differentiation in such lesions as ductal carcinoma in situ

In any case, the current case is a good example of a lesion with many benign glandular, luminal cells surrounded by more than one cell layer which can be ideptified as myoepithelial, followed by a delicate basement membrane. Thus, the majority of the cellular mass in this lesion shows myoepithelial differentiation as indicated by immunocytochemistry for actin. Although this particular tumor shows mostly the lobulated pattern as defined by Tavassoli 3 , there are also spindled areas . This is again an indication of the heterogeneity of so many breast lesions. There is probably some

4overlap with lesions which have been termed,

"Tubular adenoma" .

These mixed lesions of special variety s, 6 have been regularly benign with very rare exceptions. One particularly interesting one was a lesion somewhat similar to the present one which evolved over several recurfences to a neoplasm indistinguishable from a leiomyosarcoma .

The lesion recently designated "ductal adenoma" , may have many myoepithelial-type cells. It is a lesion which blends features of fibrosed papilloma and nodular adenosis . The major importance of ductal adenoma is that it may be misinterpreted as W~t~gnant because of peripheral pseudoinvasion and atypical nuclei

?

REFERENCES

1. Rudland PS: Histochemical organization and cellular composition of ductal buds in developing human breast: Evidence of cytochemical intermediates between epithelial and myoepithelial cells. J .Hi stochem.Cytochern. 39 : 1471- 1484, 1991.

2. Bussolat i G,Botta G and Gugliotta P: Actin - rich (myoepithelial) cells in ductal carcinoma in situ of the breast. Virchows Arch.[B] 34:251-259, 1980.

3. Tavassoli FA: Myoepithelial l esions of the breast: Myoepitheliosis, adenomyoepithelioma, and myoepithelial carcinoma. Am. J.Surg.Pa thol. 15:554-568, 1991 .

. 4. O'Hara MF and Page DL: Adenomas of the breast and ectopic

breast under lactational influences . Hum Pathol 16:707-712, 1985.

5. Rosen PP: Adenomyoepithelioma of the Breast. Hum Pathol 18:1232-1237, 1987.

6. Zarbo RJ and Oberman HA: Cellular adenomyoepithelioma of the breast. Am J Surg Pathol 7 :863-870, 1983.

7. Cameron HM, Hamperl H and warambo W: Leiomyosarcoma o f the breast originating from myot helium (myoepithelium). J Pathol 114 :89-92, 1974.

8. Azzopardi JG and Salm R: Ductal adenoma of the breast:a lesion which can mimic carcinoma. J.Pathol. 144:11-23, 1984.

9. Lammie GA and Millis RR: Ductal adenoma of the breast--A review of fifteen cases. Hum .Pathol. 20:903- 908, 1989.

10. Page, D.L. and Anderson, T.J. Papilloma and related lesions. In: Diagnostic Histopathology of the Breast, Edinburgh: Churchill Livingston, 1987, p . 104-119.

3

CASES 3 AND 4

Diaqnosis: Giant f ibroadenoma v i th focal be.niqn phyllodes tUIIIor

This very large tumor of stromal and epithelial elements from a young woman demonstrates the difficulty of terminology in these lesions. The terms giant, juvenile, and cellular have been suggested for the large and more cellular examples of fibroadenoma. Much of this attempt has been to avoid the use of the term "phyllodes tumor", or worse. The less acceptable term is the outmoded term which has survive~ for almost a hundred and fifty years, "cystosarcoma phyllodes " •

The careful gross examination of this neoplasm at the cutting bench, a mandatory exercise in these lesions, demonstrated centrally an area of a leaf-like (phyllodal) growth which was surrounded by large nodules of. more solid tumor.

The histologic pattern of the great majority of the lesion fell well within the range of that of fibroadenoma, with a modestly cellular stroma and regular ly distributed epithelial elements. The epithelial elements frequently demonstrated a gynecomastoid type of hyperplasia which is common in these lesions .

The _more central part, the leaf-like growth pattern which occasioned the original naming of these lesions as leaf-like (phyllodes) , demonstrated broad leaf-like areas histologically with a very cellular stroma immediate ly underneath the epithelium. It is evident that the latter pattern would define quite nicely what we expect histologically to be seen in most benign phyllodes tumors, and that it is the unusual cellularity of the stoma immediately related to the epithelium, as well as the very cellular stroma and grossly evident intracanalicular pattern, which make the phyllodes designation appropriate.

We are thus given a wide choice of possibilities for diagnosis in this case. Consideri ng the frequent misunderstanding of the phyllodes tumor, it is often the case that the surgical pathologist has to shape · the diagnosis to fit the proper therapy and the needs of the clinical situation. My own preference for this case would be to call it a giant fibroadenoma with focal pattern of benign phyllodes tumor . Certainly, "giant fibroadenoma" alone would be acceptable also.

Determination of malignancy in these lesions is much more important than determining which of the terminologic possibilities for the benign lesions is appropriate. There is no proof that any of the benign lesions has a greater likelihood of local recurre.nce than others 3.

The determination of malignancy has been approached quite frequently in the rece~t past beginning with a classic paper by Pietruszka and Barnes . They looked at cellularity, degree of nuclear atypia and mitotic counts as indicators for malignant behavior. There was a fair association with these, and when one looks at cases with mitosis of over 10 per 10 high power field,

4

about a quarter of the lesions may be found to be malignant as determined by distant metastasis. Much more spec ific in indicating a high likelihood of malignant behavior is the finding o f heterologous sarcomat ou s elements: liposarcoma, rhabdomyosarcoma, chondrosarco~a or osteosarcoma and extensive necrosis within the lesions 5- • Overgrowth of s troma, losing ~e~ular relation to epithelium, is also a feature of malignancy

' • Irregular borders have sometimes been associated with a greater likelihood of local recurrence, but this has not been evaluated in any careful fashion with respect to the many different ways i n which borders might be evaluated (extent and degree of irregularity).

REFERENCES

1. Hajdu SI, Espinosa MH and Robbins GF : Recurrent cytosarcoma phyllodes--a clinicopathologic study of 32 cases. Cancer 38:1402-1406, 1976.

2. Adami HOG,Hakelius L,Ri msten.A and Willen R: Malignant, locally recurring cystosarcoma phyllodes in an adolescent female . Acta Chir Scand 150:93-100, 1984.

3. Pietruszka M and Barnes L: Cystosarcoma phyl lodes--a clinicopathologic analysis of 42 cases. cancer 41 :1974-1983, 1978.

4. Cohn-Cedermark G,Rutqvist LE,Rosendahl I and Silfverswrd C: Prognostic factors in cystosarcoma phyllodes: A clinicopathologic s tudy of 77 patients. Cancer 68: 2017-2022, 1991.

5. Ward RM and Evans HL: Cystosa rcoma phyl1odes. A clinicopathologic study of 26 cases. Cancer 58 :2282-2289 , 1986.

6 . Norris HJ a nd Taylor HB: Relationship of histologic features to behavior of cystosarcoma phyllodes--analysis of ninety-four cases. Cancer 22 :22-28, 1967.

7. Hart WR,Bauer RC and Oberman HA: cystosarcoma phyllodes. A clinicopathologic study of twenty-six hypercellu1ar periductal stromal tumours of the breast . Am J Clin Pathol 70:211-216, 1978.

8. McDivitt RW,Urban JA and Farrow JHi Cystosarcoma phyllodes. Johns Hopkins Me d J 120:33-45 , 1967.

9. Pignatelli M,Hanby AM and s tamp GWH: Low expression of A1 , A2 and A3 subunits of VLA integrins in malignant mammary tumours. J.Pathol . 165:25-32, 1991.

5

CASE 5

DIAGNOSIS: Pleomorphic i n f i l t rating lobular c arcinoma wi th apocrine cytology.

This particular carcinoma is important because of its unusual pattern. Some of the infiltrating carcinoma cells could be mistaken for histiocytes. There are broad areas where the cells attach to each other, forming a solid sheet of cells with abundant finely granular, eosinophilic cytoplasm. The nuclei demons trate extreme pleomorphism. In many other areas the cells adopt the singl e cell infiltrating pattern characteristic of infiltrating lobular carcinoma.

This, then, is an example of a mixed invasive lobular carcinoma with the solid and pl~omorphic variants presenting in a solitary mass. This could also be regarded simply as a pleomorphi c varia nt, and that may be more appropriate, since the extreme nuclear atypia should indicate a bad prognosis despit e the low growth fraction .

There are several variants of invas;tve lobular carcinoma (ILC) which are now quite widel y accepted - 4 • This recogn ition of subtypes or variant; of infiltrating lobular carcinoma began with Fuhner in 1975 fostered by the purist approach to diagnosing the classic type as clearly presented by Wheeler and Enterline 6 . The most i mportant point of this discussion is that classic, invasive lobular carcinoma with regular nuclear pattern and i ndividual cell infiltration is a very special type of breast cancer. It is associated with frequently inapparent definition of the mass , making detection difficult by mammography and physical examination. The growth fraction is very low, but that most of these "special type" breast cancers demonstrate metastatic potential, but not until ten years after detection. Mammogr;phic images of infiltrating lobular carcinoma may be quite subtle .

6

REFERENCES

1. Page, D.L., Anderson, T.J. and Sakamoto, G. Infiltrating carcinoma: maJor histological types . In : Dia gnostic Histology of the Breast, edited by Page, D. L. and Anderson, T.J. Edinburgh : Churchill Livingstone, 1987 , p . 193-235.

2. Dixon JM,Anderson TJ, Page DL,Lee D and Duffy sw: Infiltrating lobular carcinoma of the breast. Histopathol. 6:149-161, 1982.

3. du Toit RS,Locker AP,Ellis IO,Elston CW,Nicholson RI and Blamey RW: Invasive lobular carcinoma of the breast-the prognosis of histopathological subtypes. Br J cancer 60:605-609, 1989.

4. Dicostanzo D,Rosen PP,Gareen ! ,Franklin s and Lesser M: Prognosis in infiltrating lob).llar carcinoma : An analysis of "classical" and variant tumors. Am J surg Pathol 14:12-23, 1990.

5. Fechne r RE: Histologic variants of infiltrating lobular carcinoma of the breast. Hum Pathol 6:3 73-378, 1975.

6. Wheeler JE a nd Enterline HT: Lobular carcinoma of the breast in situ and infiltra ting. Pathol Annu llpt.2:161-188, 1976 .

7 . Hilleren DJ,Andersson IT,Lindholm K and Linne ll FS: Invasive lobular carcinoma: Mammographic findings in a 10- year experience. Radiology 178 :149-154, 1991.

7

CASE 6

Diagnosis: Non-comedo ductal carcinoma in- situ, micropapillary type.

This is an extremely difficult case exemplifying the importance of extensiveness of disease in evaluating the clinical importance of the ductal pat\ern epithelial proli~erations . The papers of Patchefsky et al. and Lagios et al . have given us the greatest amount of information with regard to relevance of extensiveness of ductal carcinoma in situ {DCIS) as well as the predictability of success following planned wide local excision of these lesions. Both papers either imply or state directly that the micropapillary pattern, particularly when present in pure form, may be more likely to be very extensive than any other pattern, possibly exclusive of ~he well-developed comedo pattern.

This case is particularly difficult because the micropapillary pattern is not well-developed in many sites , but there are several spaces in each slide in which characteristic bulbous projections with evenly placed cells are present, and the cytologic features present within each membrane-bound space are totally uniform in an occasional space and almost completely so in many. The other interesting feature is that here, unlike so of the other types of ducdtal carcinoma in situ, large ducts are primarily involved, with lesser involvement of lobules with the so-called cancerization pattern 3 •

In summary , this case is diagnosable as micropapillary ductal carcinoma in situ of a lower grade than usually seen. The major implication is that it may be quite extensive within the breast and that if local excision is to be attempted following the initial biopsy, care must be given to planning 4

REFERENCES

1. Patchefsky AS,Schwartz GF,Finkelstein SD,et al: Heterogeneity of intraductal carcinoma of the breast. Cancer 64: 731-741, 1989.

2. Lagios MD,Margolin FR,Westdahl PR and Rose MR: Mammographically detected duct carcinoma in situ . Cancer 63:618-624, 1989.

3. Azzopardi , J.G . Problems In Breast Pathology, Philadelphia:W. B. saunders, 1979. pp. 203-213.

4 . Lagios MD: Duct carcinoma in Situ: Pathology and treatment. Surg.Clin . N.A. 70 : 853-871, 1990 .

8

CASE 7

Diagnosis: Adenoid cystic carcinoma

This rare type of invasive breast cancer has some uncertain confines of definition. However, it also has a classic clustering of features which when present in a consistent fashion are characteristic and have a very important prognostic association.

The sharply circumscribed, virtually perfect geometric circles containing mucoid and/or baseme.nt membra.ne material within sharply defined islands of cells are characteristic of this tumor. Nodal metastases are highly unusual and estrogen and progesterone receptors are routinely negativ~ .

A somewhat inconsistent feature of this lesion which continues to be mildly bothersome is that its prognostic implication is precisely the opposite of that of the similarly named lesion of the major salivary glands. That is to say, these lesions are expected to have a clinical evolution when appeiring in the breast identical to that of invasive tubular or cribriform carcinoma. The latter is discuss~d below, and bears close histologic rsemblance to adenoid cystic • Unless there are unusual presentations, women with these completely e.xcellent prognostic carcinomas can usually be assured that recurrence is not a realistic possibility.

Invasive cribriform carcinoma was first defined and presented in the literature in 1983. There is one recent follow-~p paper on a series of patients from George Washington University which has largely corroborated the lack of later distant metastases with these tumors. Local recurrence has been described, but distant metastases have not. Invasive cribriform carcinoma is defined as an invasive carcinoma in which the invasive component has the appearance of cribriform carcinoma in-situ. Adenoid cystic carcinoma has not only mucoid material in shatt?ll

5defined, round

spaces, but also basement membrane material ' ' • Lymph nodal metastases are occasional·ly present as in tubular carcinoma, have been in fewer than 3 lymph nodes, and have not been associated with distant metastatic disease.

REFERENCES

9

1. Baker RR: Unusual lesions and their management. Surg.Clin.N.A. 70:963 -975 , 1990.

2. We lls CA,Nicoll S and Ferguson DJP: Adenoid cystic carcinoma of the breast: a case wi th axillary lymph node metastasis. Histopathology 10: 415-424, 1986.

3 . Venable JG, Schwartz AM and Silverberg SG : Infi ltrating cribriform carcinoma of th e breast: A di s tinct i ve c linicopatho logic entity. Hum . Pathol. 21:333-338, 1990.

4. Peters GN and Wolff M: Adenoid cystic carcinoma of the breast. Report of 11 new cases: review of the literature and d iscussion of biological behavior. Cancer 52:680-686, 1982.

5. Zaloudek c,oertel YC and Orenstein JM: Adenoid cystic carcinoma of the breast. Am J Clin Pathol 81:297-307, 1984.

10

Dia g nosis : non-invas ive carcinoma.

CASE 8

Multip le duct papillomas v i th focal e ncy sted, papillar y carcinoma and adjacent i n- sit u ductal

Although the elongated diagnosis rendered in this case may be more complex than the case itself, it does document the various elements present. · What is most important is that there is a high likelihood of local recurrence if a mastectomy. has not been performed in a male. It is adequate wide excision which is important in these lesions, and they should be regarded as similar to ductal carcinoma in-situ, which is cured by adequate wide excision

1 ~ut care must be given to evaluate the extensiveness of

disease • •

This case must be regarded as analogous to the encyst~d papillary carcinoma paper of Carter et al published in 1983 • They pointed out the importance of adjacent ductal carcinoma in-situ in these lesions, which were easily cured without regional recurrence unless adjacent ductal carcinoma in-situ was present.

This case is actually not a simple,, solitary , encysted, non-invasive papilary carcinoma, but really an example of multiple duct spaces involved by papillary proliferations which focally meet the criteria of ductal carcinoma in-situ. It is likely that ~iven the wide acceptance of the applicability of local excision

, that one should err on the side of overdiagnosing non-invasive carcinoma in this setting in order to assure that the recurrence rate is minimized . This case docume.nts the incredible variety of patterns present in these cases, and the frequent difficulty of certain diagnosis. I would like to add that ]he necessity for careful sampling of these lesions is mandatory as not only may adjacent ductal carcinoma in-situ be found, but associated areas of invasive carcinoma may be co-existent as well.

Approximately five percent of male carcinomas are of the non-invasive papillary type with a further 5% being in-situ duct carcinoma 6 . Because of the rarity of male breast carcinoma in general, the occurrence· of these patterns frequently presents consernation to clinicans who don't have a great experience in breast disease.

It would seem that the performance of a formal mastectomy, with perhaps a close shaving of the nipple (as it may not be necessary to remove it) should be done in most cases in which a mass is present in the male breast which does not have the gross appearance of gynecomastia .

11

References 1. Andersen JA,Neilsen M and Bilchert-Toft : The growth pattern of in situ carcinoma in the female breast. Acta Oncol 27:739 - 74 3, 1988.

2. Gump FE,Jicha DL and ozello L: Ductal carcinoma in situ (DCIS) : A revised concept. Surgery 102:790- 795, 1987 .

3. Carter D,Orr SL and Merino MJ : Intr acystic papillary carcinoma of the breast . After mastectomy,radiotherapy or excisiona l biopsy alone. Cancer 52:14-19, 1983.

4. Lagios MD: Duct carcinoma in Situ: Pathology and treatment. Surg.Clin.N.A. 70 : 853-871, 1990 .

. 5. Ohuchi N,Rikiya A and Kasai M: Possible c ancerous change of intraductal papillomas of the breast: a 3D reconstruction s tudy of 25 cases. Cancer 54:605-611, 1984.

6 . Heller KS, Rosen PP, Sch ottenfeld D and et a l : Male breast cancer: a clinicopathologic study of 97 cases. Ann Surg 188 : 60-65, 1978.

12

CASE 9

Diagnosis : Radi a l Scar (complex Sclerosing Lesion)

This well-developed radial scar or complex sclerosing lesion may be somewhat unusual in this young age group ('20's). I n any case, the biopsy is quite normal with the exception of a well-formed dense rounded scar surrounded by a radial array of epithelial elements . The scar is quite classic but with less elastic tissue than is often seen and containing benign epithelium entrapped within the scar. The epithelial elements surrounding the scar demonstrate quite extensive examples of florid epithelial hyperplasia without atypia . These patterns are sometimes alarming because of the compactness of the cells, particularly so when histologic sections are thicker than usual. This makes the narrow slit-like spaces characteristic of benignancy less apparent, and makes the case look even more cellular and seemingly comprised of cells with a greater nuclear-cytoplasm ratio.

Radial scar is the term given to us by the radiologists who do mammography. It is a term mandated by finding these radial arrays whic mimic carcinomas on x - ray 1-~. They are regularly benign and have no proven favored association with carcinoma. However, small carcinomas may be present adjacent to them which are usually of tubular type. The term "complex sclerosing lesion" has been proposed for the larger and more complex lesions which particularly mimi c carcinoma with the term implying that there is greater complexity and therefore more likelihood of mimiking carc inoma 4 •

REFERENCES

1. Wellings SR and Alpers CE: Subgross Pathologic Features and Incidence of Radial Sca~s in the Breast . Hum Pathol 15:475-479, 1984.

2. Adler DD,Helvie MA,Oberman HA,lkeda OM and Bhan AO: Radial sclerosing lesion of the breast: Mammographic features. Radiology 176:737-740, 1990.

3. Battersby S and Anderson TJ: Myofibroblast activity of radial scars. J.Pathol. 147:33- 40, 1985.

4. Page, D.L. and Anderson, T.J . Radial Scars and Complex Sclerosing Lesions. In: Diagnostic Histopathology of the Breast, edited by Page, O.L. and Anderson, T . J . Edinburgh: Churchill Livingstone, 1988, p. 89-103.

§)U u

13

CASE 10

Diagnosis: Amyloid Deposits

This biopsy is an excellent example of long-term amyloid deposits characterized by the presence of foreign body giant cells. Amyloid deposits in the breast have been rarely described and are usually associated with systemic amyloidosis which may be

_ either of the light chain, immunoglobulin type or of the true secondary amyloidosis seen with such diseases as rheumatoid arthritis 1 ' 2 • Large masses of amyloid, as seen in this case, as opposed to more subtle deposits largely in a perivascular location, have been associated more with true secondary ~m~loidosis than the type associated with monoclonal gammopathy

- • Usually, the clinical setting in which amyloid deposits are found will indicate the type of amyloidosis.

REFERENCES

1 . Glenner G and Page D: Amyloid amyloidosis and amyloidogenesis . Int Rev Exp Pathol 15:1-92, 1976.

2. Glenner GG: Amyloid deposits and amyloidosis (Part 1) . N Engl J Med 302:1283-1291, 1980.

3. Cetti R,Reuther K,Hansen J and Schiodt T : Amyloid tumor of the breast. Dan Med Bull 30 :34-35, 1983.

4. Hecht AH, Tan A and Shen JF: case report: Primary systemic amyloidosis presenting as breast masses, mammographically simulating carcinoma. Clin.Radiol. 44:123-124, 1991.

5. Sadeghee SA and Moore SW: Rheumatoid arthritis, bilateral amyloid tumors of the breast and rnul tiple cutaneous amyloid nodules. Am.J.Clin.Pathol. 62:472-476, 1974 .

§' -

14

CASE 11

Diaqnosis: Hyxoid Liposarcoma

This case presents difficulties only because of the fact that t he presence of this lesion within the breast is unusual. If the site had been thigh or retroperitoneum, the diagnosis would have been relatively easy. Myxoid liposarcomas are regularly present in the sites noted above, and are much less frequent within the breast. However, of the primary sarcomas presenting within the breast, liposarcoma is relatively common and this varietY. of liposarcoma is the one most frequently seen within the breast 1 • 2 .

Histologically, the characterist ic plexiform n e twork of capillary vessels is seen in only a small proportion of this lesion. Much of this lesion is made up of pools of the lightly staining myxoid material. Both larger pools and extensive lacelik e areas are present 3 • The lesion is well demarcated. Lesions such a s thi s are regularly cured by adequate wide local excision which may not entail a complete mastectomy.

REFERENCES

-·

1. Austin RM and Dupree WB: Liposarcoma of the breast: a cl inicopathologic study of 20 cases. Hum Pathol 17:906- 913, 1986 .

2. Kristensen PB and Kryger H: Liposarcoma of the breast. Acta Chir Scand 144 : 193-196, 1978.

3. Enzinger, P.M. and Wei ss, s.w. Liposarcoma . In: Soft Tissue Tumors , edited by Stamathis, G. St. Louis : C.V. Mosby Co., 1988, p. 346-382 .

15

Case 12

Diagnosis: Metaplast i c car cino•a , c hondrosarco•a

This is an excellent example of metaplastic carcinoma of the breast in which cartilaginous differentiation is well- developed. It is not always the case that epithelial markers are easily demonstrated in these lesions by immunocytochemistry, but in this particular case epithelial membrane antigen and cytoke ratin antigen determinants were easily demonstrated by immunocytochemistry with antibodies to these substances.

There has been continuing controversy about the behavior of these unusual lesions . In geperal, it would seem that those presenting with only sarcomatous differentiation will behave as sarcomas, and should be treated as such. That is to say that unless they have an extensive epithelial component they will not metastasize to the lymph nodes .

In broad terms, whatever terms are chosen, these lesions of mixed epithelial and mesenchymal differentiation should be diagnosed with clarity wi th inclusion of the percentage and precise histologic characterization of the components.

One large series published in 11_84 from the M.D. Anderson Hospital is a hallmark presentation and has been augmente~ 2Y Wargotz et al from the Armed Forces Institute of Pathology - . The latter have separated the metaplastic carcinomas into three broad group~; matrix producing, spindle cell and carcinosarcoma (CS) . The latter category is recognized if intraductal or infiltrating carcinoma ofthe breast is contiguous or subtly merged with a highly cellular, mitotically active pleomorphic spindle cell carcinoma. This group forms the more orderly spindle cell carcinoma pattern.

Matrix-producing metaplastic carcinoams were defined by Wargotz et al as those tumors having transition from overt carcinoma to chondroid andjor osse~us stromal matrix withou t an intervening spindle cell component • These latter neoplasms had an overall 10 year survival of 68% as compared to 40% for the carcinosarcoma patients 5, Limited size and stage are associated with a better prognosis. The high histologic grade of the CS cases is reflected in their average mitotic count of 19 mitoses for 10 high power fields as compared to an average of three for the carcinomas with spindle ce~ metaplasia. Of the 70 women with cs studied by Wargotz et al , 26% had metastases to axillary lymph nodes, predominantl y of the carcinomatous component.

Comparison of the cs cases with those of spindle cell carcinoma revealed a cumulative five year survival of the spindle cell carcinoma patients of 64% as compared to 49% for cs patients . Note that with a 10 year survival of 40% for the cs patients, late death from tumor is less likely than early death .

16

Metaplastic carcinomas are a diverse group of tumors, and as with so many other lesions of the breast, await a satisfactory resoluti on of the dilemma of the lumpers and the splitters. It may well be that grading the stromal component as is done for soft tissue sarcomas, would produce a satisfactory stratification of these l esions for prognostic purposes. For example, phyllodes tumors when malignant may well be most satisfactoral ly stratified for prognostication by grading the malignant stromal component and estimating its size . Such suggestions will await study of combined groups of patients, a goal perhaps unattainable in such rare tumors .

In the majority of the cs and spindle cell cases, at least some of the stroma may be decorated with antibodies for keratins. While this is taken as strong evidence for the metaplastic character of this mesenchymal c9mponent, it may not be necessary to render this test. The rationale for this approach is that in the absence of keratin being demonstrated in the stromal component, the diagnosis will remain the same as long as we use the approach of accepting as the explanation for these "mixed tumors".

In the Kaufmann et al paper 1 , prognosis was related to several definable elements but was considerably worse, controlled for clinical stage, than usual breast cancers. Larger tumors have a poorer prognosis. Most defining was the predominance of epithelial v . sarcomatous elements. Women with predominantl y epithel ial tumors had a 62% five-year actuarial survival, while those with predominantly sarcomatous malignancies had only a 28% survival experience. Obvi ously, the spindle cell tumors were not included.

The complexity of keratins demonstrated in such mesenchymal neoplasms as smooth muscle neoplasms and vimentin and desmin in seewing epithelial ne9lasms has been recently reviewed by Jones et al and Pitts et al •

17

REFERENCES 1. Kaufman MW,Maerti JR,Gallager HS and Hoehn JI: Carcinoma of the breast with pseudosarcomatous metaplasia. Cancer 53: 1908-1917, 1984 .

2. Wargotz ES and Norris HJ : Metaplastic carcinomas of the breast. I. Matrix-producing carcinoma. Hum.Pathol. 20:628-635, 1989.

3. Wargotz ES,Deos PH and Norris HJ : Metaplastic carcinomas of the breast. II. Spindle cel l carci noma. Hum.Pathol. 20:732- 740, 1989.

4 . Wilkinson EJ,Schuettke CM,Ferrier CM,Franzini DA and Bland KI: Fine needle aspiration of breast masses : An analysis of 276 aspirate.s. Acta Cytol . 33:613-619, 1989.

5. Wargotz ES and Norris HJ: Metaplastic carcinomas of the breast: III. Carcinosarcoma. cancer 64:1490-1499, 1989.

6. Jones EA,Flejou J-F,Molas G and Potet F: Pleomorphic carcinoma of the small bowel. Pathol.Res.Pract . 187:235-240, 1991.

7. Pitts WC,Rojas VA,Gaffey MJ,et al: Carcinomas with metaplasia and sarcomas of the breast. Am.J.Clin.Pathol . 95:623-632, 1991.

18

CASE 13

Diagnosis: osteosarcoma

The lesion appears to be a histologically pure osteogenic sarcoma. With no evidence to the contrary, and lacki ng the varied patte.rns seen in mosr <Fetapl astic carcinomas, this should be regarded as a sarcoma , .

References 1. Savage AP,Sagor GR and Dovey P : Osteosarcoma of the breast : a case report with an unusual diagnostic feature. Cl in On col 10:295-298, 1984.

2. Going JJ,Lumsden AB and Anderson TJ: A classical osteogenic sarcoma of the breast: histology, immunohistochemistry and ultrastructure. Histopathology 10:631-642, 1986 .

19

Case 14

Diagnosis: Florid hyperplasia with cysts

These slides represent an excellent example of relatively circumscribed changes of what has long been called "fibrocystic disease" in a young woman. Stewart and Foote called attention to the frequent clustering of several changes: cysts, stromal formation

1 epithelial hyperplasia, apocrine change and sclerosing

adenosis • There has been a recent attempt to remove the word "disease"

from these representations, and replace it with something like "fibrocystic change". Considering that there is no indication that cysts by themselves are associated in any clinically meaningful way with pain, increased risk of cancer, and are present in a majority of preimenopausal women, it is difficult to believe that they should receive the designation of "disease". It is possible that larger cysts, occasioning asp~ration, may be regularly accompanied with epithelial hyperplasia which h~s been associated with predictability of later cancer development - 5 but these matters are still under study.

The apprance of florid hyperplasia, cysts and other changes in young women has been termed "juvenile papillomatosis". The precise clinical implications of this are still not clear as the association with carc\Jloma has not reached statistically signifiance in onestudy . One of the reasons is that studying young women for increased cancer incidence demands very · long follow-up until they get into the age groups in which carcinoma usually occurs.

REFERENCES

1 . Foote FW and Stewart FW : comparative studies of cancerous versus noncancerous breasts. Ann Surg 121:6-53;-197-222, 1945 .

2. Bundred NJ,West RR,Dowd JO, Mansel RE and Hughes LE: Is there an increased risk of breast cancer in women who have had a breast cyst aspirated. Br. J.Cancer 64:953-955, 1991.

3. Page DL,Dupont WD,Rogers LW lesions of the female breast. 55:2698- 2708, 1985.

and Rados MS: Atypical hyperplastic A long-term follow-up study. cancer

4. Dupont WD and Page DL: Risk factors for breast cancer in women with proliferative breast disea.se. N.Engl.J.Med. 312:146-151, 1985.

5. Page DL and Dupont WD: Anatomic markers ot human premalignancy and risk of breast cancer. Cancer 66:1326-1335, 1990.

6 . Rosen PP and Kimmel M: Juvenile papillomatosis of the breast: A fo llow-up study of 41 patients having biopsies before 1979. Am.J.Clin.Pathol. 93:599-603, 1990 .

?n

Case 15

Diagnosis: "Giant cell" carcinoma or anaplastic carcinoma

This lesion does not have tumoral giant cells, but is clearly a carcinoma. There are some focal features of medullary carcinoma, but that certainly would be an incorrect diagnosis with the many areas of invasion into the stroma present around the edges of the tumor. It is somewhat surprising that there are not as many mitotic figures as are frequently present in carcinomas having the other features of this lesion . One can only state that being certain that it is a carcinoma, and very likely to be a primary, it should be treated with a great amount of respect. This is clearly a high grade carcinoma 1, 2 .

Most high grade tumors which present as metastases within the breast from another site are small cell carcinomas of the lung of the intermediate or large cell variety. Most of the metastatic lesions that present within the j>reast are with the clinical knowledge of the primary elsewhere •

1. Elston cw and Ellis IO: Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathol. 19:403-410, 1991.

2. Henson DE: The histological grading of neoplasms. Arch Pathol ~ Med 112 : 1091-1096, 1988.

3. Di Bonito L,Luchi M,Giarelli L,Falconieri G and Viehl P: Metastatic tumors to the female breast: An autopsy study of 12 cases. Pathol.Res.Pract . 187:432-436, 1991.

21

CASE 16

Diagnosis: Pag et's Disease of the nippl e

This case may be recognized as a classic presentation of Paget's disease of the nipple. Quite obvious, large, usually round malignant cells with a tendency to vesicular nuclei and large nucle.oli are interspersed within the epidermal cells of the· nipple. The disease can spread from the nipple to the areola and even to surrounding skin, but such extensive disease is rarely seen in current practice.

Recent experience documents that approximately five to ten percent of cases presenting as c.linically symptomatic Paget's disease may have disease confined to the nipple or to the immediately underlying ductal spaces. This is stated only to highlight the fact that disegse is commonly fairly extensive within the breast. I believe that much of the time the appearance of Paget's disease in the nipple is an indicator of extensive ductal carcinoma in- situ in all four quadrants of the breast . In about fifty percent of cases, (reduced perh,aps to 30% in the mammographic era) there are palpable invasive carcinomas within the breast as well at the time of diagnosis. It is evident then that in the presence of Paget's disease of the nipple, the first necessity is to determine the extent of the disease within the breast. The 50% figure of haying a dominant invasive carcinoma is supported by Salvidor et al . . The experience of the breast unit at Guy's Hospital also supports the idea that Paget's disease ~f the nipple usually includes extensive disease within the breast , and further demonstrates the immunohistochemical characterization of the Paget cells mirrored that of the underlying in- situ carcinoma. Immunohistochemistry for cytolceratin, which is usually only lightly positive if at all in the epidermis, and epithelial membrane antigen, as well as the various human fat milk globule antigens will decorate the Paget's cells strongly in the great majority of cases . These may not be necessary in the presence of Alcian blue-paS positivity . However, there remain some cases in which immunohistochemistry is really necessary in order to be certain that Paget's disease is present, particularly when the diagnosis is subtle with few cells present.

The fact that the disease may be relatively corfined within the nipple has . been highlighted by Lagios et al. • They and others have had a successful experience with wide excision of the nipple-arelolar complex in Paget's disease when confined to the nipple . However, when one takes this approach it is quite obvious that margins need to be .evaluated carefully . This is quite easy to do as the ducts are all at one face of the lesion the majority of the margins are outer skin.

References

22

1. Salvadori B,Fariseli G and Saccozzi R: Analysis of 100 cases of Paget's disease of the breast. Tumori 62:529 - 536, 1976 .

2. Chaudary MA,Millis RR,Lane EB and Miller NA: Paget's disease of the nipple: a ten year review including clinical, pathological, and immuno-histochemical findings . Breast Cancer Res Treat 8:139-146, 1986.

3. Lagios MD,Westdahl PR,Rose MR and Concannon s: Paget's disease of the nipple:alternative management in cases without or with minimal extent of underlying -breast carcinoma . Cancer 54 : 545- 551, 1984.

23

CASES 17 AND 18

Diagnosis: Bypersecretory hyperplasia, cysts and xanthomatous duct ectasia.

This case is most unusual, particularly in the presentation of so much enlargement of the breast . The major change in the initial biopsy appears to be the dilatation of the interlobular ducts and the extra-lobular terminal ducts dominantly by a milk-like material. There are many alterations of the breast which mimic lactational change in some way, none of which are well-understood. Occasional individual lobular units which look precisely correct for lactational change have been described since the 1920's. When many of these are present it may indicate an elevation of prolactin, which usually turns out to he secondary to medication .

More e.xtensive changes have .been described by Rosen as cystic hypersecretory hyperplasia 1 • 2 , but this case may not so qualify because the cyst formation is not as apparent . In any case, the changes in this particular instance are certainly benign.

The second biopsy has v ery prominent f oamy macrophages or foamy epithelial cells. The foam cells found in the ducts along with lactation are often epithelial cells: when seen in other s i tuations they are often macrophages . In this instance these may well be either, as it seems to be a late change dependent upon the presence of the milk-like material. Some of the appearances are quite alarming in that one finds large areas of foamy cells present in the luminal location as well as under the epithelium. Broad areas of disruption in the epithelium are a sign that this is a secondary alteration.

REFERENCES

1. Guerry P, Erlandson RA and Rosen PP: Cystic hypersecretory hyperplasia and cystic hyerpsecretory duct carcinoma of the breast: Pathology, therapy, and followup of 39 patients . Cancer 61:1611-1620, 1988.

2. Jensen RA and Page OL: Cystic Hypersecretory carcinoma: What's in a name? Arch Pathol Lab Med 112:1176, 1988.

24

CASE 19

Diagnosis: High grade, largely encysted papillary c arcinoma , microinvasiva.

Although this case is undeniably an encysted or at least wel l-defined papillary carcinoma, i t is not the usual pattern seen in the breast.

The classic paper of carter et ~1. 1 of almo~t 10 years ago established the "encysted" , non-invasive papillary carcinoma of the breast as a special entity . With a relatively small number of cases, but clear outcomes with follow-up, the encysted lesions without adjacent duc t a l carcinoma in situ (DCIS) were shown to be curable by local excision. Cases with ocrs outside the lesion then become similar to the more recent concept of extensive in situ carcinoma of ductal type :which often recurs after a local excision. The therapeutic implication of extensiveness of ocrs is quite analogous with or without an encysted papillary lesion being present. With regard to the current case, however, the histologic pattern of the non-invasive component itself has two major differences from any of the patterns usually seen in the lesions described by Carter, et al. Specifically this material looks like a carcinoma of the colon, and is highly cytologically atypical, rather than having the patterns of ductal carcinoma in situ within the encysted lesion. Also, the interface between the tumor and the surrounding tissues is not as sharp as it usually is. Instead , there is a fine irregularity made up of very loose connective tissue filled with chronic inflammatory cells . In the usual non- invasive papillary case the margin is so sharp as to i ndicate the presence of the same sort of basement membrane structure present along the fibrovascu l ar fronds within the lesion.

I would conclude that this case may well have metastatic potential and furthermore could be a metastasis from elsewhere in the body, parti cularly the gastrointestinal tract.

References

1. Carter D,Orr SL and Merino MJ: Intracystic papillary carcinoma of the breast. After mastectomy,radiotherapy or excisional biopsy alone. Cancer 52: 14-19, 1983.

Case 20

Diagnosis: Silicone granuloma

The diagnosis in such cases is relatively easy when the history is known. The whole story of the reaction of the soft tissues of the breast to silicone in its various forms is an unfinished book.

The histologic pattern is really quite different from ordinary fat necrosis, although the differences are somewhat subtle. In fact, in necrosis at its various stages there is regional variation which finally leads to a central area that is mostly enlarged spaces with fluid fat surrounded by spaces of varying size of "fat cysts" and fibrous scar until the fibrous scar predominates at the outer edges. Although there may be foamy macrophages present, the extreme.variation of vacuolization within cells seen with silicone, and well-demonstrated in this case, are not seen in ordinary fat necrosis.

Basically, if one wants to prove that the material present is silicone one has no special stain alternatives and one seems to be committed to electron dispersion analysis or similar special physical ie~hniques for the specific identification of these materials -

REFERENCES

1. Warner E,Lipa M,Pearson D and Weizel HA: Silicone mastopathy mimicking malignant disease of the breast in Southeast Asian patients . Can.Med.Assoc.J. 144:569-571, 1991 .

2. Thomsen JL,Christensen L,Nielsen M,et al: Histologic changes and silicone concentrations in human breast tissue surrounding silicone breast prostheses. Plast.Reconstr . Surg. 85 : 38-41, 1990.

3. Barnes L and Pietruszka M: Sarcomas clinicopathologic analysis of ten cases. 1977.

26

of the breast -- a Cancer 40:1577-1585,

In press, Seminars in Oncology

April , 1992

Prognostic Value of Histopat.hology in the Breast

~ean F. simpsorf, M.D. and David L. Page, M.D.

Department of Pathology

Vanderbilt University Medical Center

Nashville , Tn 37232

*Now at City of Ho pe Medical Center, Duarte, California

1

Introduction

Histologic preparations of breast tissue continue to be the

standard for the diagnosis of breast cancer. The further his­

tologic subclassification of breast cancers into prognostically .

meaningful groups has neither been rigorously pursued nor en-

thusiastically tested in the recent past, and is thus an un-

plumbed resource of potential prognostic importance. This review

will detail potential and proven settings in which histologic

types and/or histologic grading of breast cancers have provided

prognostic information. We wil l also review a limited number of

studies that have compared histopathologic features with nonhis-

tologic parameters.

Basically, analysis of histopathology may be by individual

characteristics such as nuclei, gland formation , etc. (grading),

or by recognizing the clustering of various cytologic and his-

to logic features into special types of carcinoma. Either of

these approaches may give prognostically important information.

This int"ormation would be most useful in recognizing women with

extremely good prognos is, i . e . , unlikely to die from breast can-

cer, or by the recognition of women with a particularly dismal

prognosis in the short term. Histopathology, like other putative

2

and proven prognostic indicators, requires careful analysis and

requisite testi ng for repetitiveness and reproducibility in order

to determine predictive worth in company with other i ndicators.

we feel that by the use of grading, we are capable of iden­

tifying a group of women (representing perhaps 20-JO% of invasive

breast cancers) who have a very poor prognosis. It is the

recognition of special types which provides certainty with regard

to excellent prognosis, with survival identified by some special

types approachi ng that of the general population as controlled

for age .

What the above implies and we will now formally state is

that a certain indicator may attain its greatest utility as an

indicator of either good or bad prognosis, and not be operative

across the entire spectrum of prediction. There has been a

widely held assumption that single indicators will be useful

across the full spectrum of tbe different stages and types of

breast cancer with regard to time course of aggressiveness,

ability to metastasize, growth fraction, etc. It will be more

useful in the immediate future for us to assume that we may need

several prognostic indicators andjor that individual indicators

may be useful only in one or another of these different spheres.

This point is also relevant to the possible restriction of pre­

dictive relevance to some stages of disease, rather than a l l.

For example, some "poor" prognostic indicators are found in TO or

3

car cinomas in situ (such as c- erb ·B-21), in which case their

presence does not mean a poor prognosis as carcinomas in situ

are incapable of metastasis.

Special Histologic Types of Invasive Carcinoma

Special types of invasive mammary carcinoma are defined by

specific histologic cr iteria which recognize a clustering of fea­

tures in some breast cancers . Through rigorous appiication of

histopathologic definitions, subsets of invasive carcinoma may be

recognized and usually indicate an excellent prognosis. Table 1

displays both general agreement as well as variation in terminol­

ogy o f special types . Their incidence is approximately 20 to 30%

of all invasive breast carcinoma.s (this figure has increased with

the mammo.grapll.i c detection of smaller carcinomas2) : the remain­

ing 70 to 80% are invasive "ductal" or "no special type (NST) " , a

term of exclusion which r einforces the practice of recognizing

and r eporting special types of invasive carcinoma. Special types

of carcinomas are defined by recognizable histologic features

present homogeneously throughout the lesion (>90% of the lesion).

When special features are present in 75-90% of the carcinoma

(variant) an improved prognosis is recognized, a lthough these

variant patterns are not regularly powerful predictors. The col­

lective utility of the special types of carcinoma as well as

their variants is illustrated by tbe survival graph shown in Fig

1 . Importantly, the survival figure was derived from patients

4

diagnosed before mammography when the average tumor size was 4

em. Thus, even women with large carcinomas, when of special

type, are unl ikely to die from breast cancer.

Tubular carcinoma is probably the most important special

type of carcinoma because of its relatively high incidence and

certainty of prediction. When present i n pure form, it is highly

unlikely to have distant metastatic potential. A recent review

of surgical therapy states that no further therapy is required in

cases of pure tubular carcinoma which are excised with a 1 to

2-cm margin of normal tissue . If the marginal tissue is free

from carcinoma , neither mastectomy, radiation, nor axillary lymph

node dissection is necessary 3 • Because tubular carcinoma repre­

sent only J to 5% of invasive carcinomas, this s pecia l type as­

sumes little significance when cases are grouped and analyzed by

stage alone . The importance of tubular carcinomas becomes ap­

parent when indivi dual patients require therapeuti c decisions.

Invasive cribriform carcinoma is very similar , both histologi­

cally and biologically t o tubular carcinoma 4

Another excellent prognosis carcinoma is mucinous (colloid)

carcinoma , def ined by t he presence o f pools of extracellular

mucin within which aggregates of low-grade tumor cells appear to

float. Mucinous carcinoma comprises approximately 2-4% of in­

vasive carcinomas and is usually present in older women. In a

Japanese study o f 175 mucinous carcinomas, 80% were present in

5

pure form. The remaining 20% showed mucinous carcinoma mixed

with areas of NST invasive c arcinoma 5 • Patients with pure

mucinous carcinomas had a 10 year survival of 90%, whereas 66% of

patients with pure mucinous carcinoma s urvived 10 years. Other

studies support and confirm this confinement of excellent prog­

nosis to histologically pure examples of this special type of

breast carcinoma 6,7

Generall y, special type carcinomas are of l ow grade (se e

below). An exception is medullary carcinoma (MC), characterized

by microscopic features of a high grade malignancy: lack of

tubule f ormation, nuclear pleomorphism, and high mitotic rate.

However, its other defining histologic elements indicate a much

better prognosis than indicated by grading alone . When the his­

tologic criteria precisely defined by Ridolfi e t al 8 are

rigorously applied, 92% of patients experience a 10 year

disease-free interval 9 e ven if the tumor is large and there are

involved lymph nodes. A study by Wargotz and Silverberg 10 was

in essentia l agreement, showing a five year survival of 95% . In

addition, these authors found that the presence of focal micro­

scopic invasion beyond the smooth borders of the carcinoma, a

sparse mononuclear cell infiltrate, or in-situ carcinoma did not

lessen the good prognosis of these lesions. If more than one of

these feat ures was present, however, a good prognosis was less

likely. These latter features have been used to relegate cases

to the atypical medullary carcinoma (AMC) 8 with a prognosis ·in-

termediate between pure MC and carcinomas of no special type . A

large study of patients from eight different NSABP protocols

showed an increased 10 year survival in patients with medullacy

carcinoma and negative lymph nodes who did not receive adjuvant

chemotherapy and in those with positive nodes who were treated

from one protocol 11 The authors concluded that medullary car--

cinoma had a better prognosis than carcinoma of no special type

but questioned whether it had a "good" prognosis . This dif-

ference from the above noted studies may be explained by the in-

elusion of some cases which others would not, by strictest

definition, diagnose as pure MC. Indeed, the NSABP authors show

an incidence of 5% which is higher than other studies S-lO.

These studies identify some of the controversy associated with

diagnosis of special types of carcinoma, and reinforce the man-

date for rigorous application of proven histopathologic criteria

in identifying prognostically useful information. It may well be

that the importance of the medullary features is greatest in

node-negative patients indicating a much better prognosis than

the mimicker, highest grade, non-medullary invasive cancer. This

notion has not been carefully tested .

Invasive lobular carcinoma is associated with special clinical

associations as well as prognostic information. Its identifica­

tion both by physical examination and by mammography 12 is ham­

pered by i ts frequent diffuse and discontinuous growth pattern

within the breast. With strict assessment of cytologic and pat-

7

tern criteria, 13- 15 , lobular carcinoma is a particularly useful

special type of breast carcinoma. Women with pure, classical type

of invasive lobular carcinoma have an excellent short term prog­

nosis, with about 90% surviving for ten years, almost without

regard to stage . Note that this rigorous definition involves

cytologic and tissue pattern criteria 16 • However, treatment

failure is frequent after that time, with a greater likelihood of

metastasis developing, often involving the gastrointestinal

tract. Prognostic inform.ation about local recurrence of in-

filtrating lobular carcinoma has been provided by Schnitt et al

17 who have shown that after local excision and radiotherapy, in­

filtrating lobular carcinoma has a local recurrence rate at five

years intermediate between NST carcinomas without (5%) and those

with (23%) extensive in-situ component of ductal type .

Possible associations between breast cancer risk factors and

special types of breast carcinoma were studied as part of the WHO

Collaborative study of Neoplasia and Steroid Contraceptives 18

Lobular and tubular carcinomas occurred with increased relative

frequency in most high risk groups determined geographically.

The proportion of these special types was also significantly in­

creased in those with bilateral breast cancer, and in those whose

first birth occurred at a later age. There was no association

with family history of breast cancer.

8

Grading of Mammary Carcinomas

currently, careful grading of breast cancer is infrequently

and imprecisely pursued despite its frequent and mandated use at

other sites such as the prostate 19, 20. In the 70-80% of in-

vasive mammary carcinomas remaining after recognition of special .

types, important prognostic information may be attained by care-

ful histologic grading. The combined histologic grade (modified

Bloom and Richardson grading system) 21 consi sts of assessment of

three features: tubular formation, nuclei, and mitoses (Fig 2).

This system fosters interobserver agreement 22-24. The Nottin--

gham prognostic index, obtained using a combination of tumor

size, nodal involvement, and combined histologic grade has been

shown to be a reproducible method for estimation of prognosis

(Fig 3) 25 . Note that this group has also tried other methods of

prognostication such as ploidy and hormone receptors, but none

has added significantly to the model used here with the three

classic predictors .

Nuclear grading has been championed by Black 26 and Fisher

27 although the results are difficult to reproduce in this single

component of the combined histologic grade 28 • McGuire et al 29

reviewed prognostic markers and supported nuclear grading. Few

studies have separately analyzed the three components of the com-

bined histologic grade, which include tubular formation, nuclear

pleomorphism, and mitotic rate. Parl and Dupont 30 found that

9

tubule formation was the strongest indicator of good prognosis

while the presence of mitoses was the greatest indicator of a

poor prognosis. A high mitotic count, although imperfectly

correlated with more precise elements which measure cell turn­

over, is an excellent indicator of short term prognostic failure

31, 32 On the other hand, the absence of mitoses is not a reli­

able indicator of good prognosis. Additional support for prog­

nostication based on mitoses is furnished by Baak 33 despite the

apparent analytic confounders 34 • It is clear that studies must

continue to clarify which of these variables have predictive

utility as well as reproducibility. The chief advantage of prog­

nostication from histologic grade is that powerful information

may be obtained from the routinely availabl e histologic prepara­

tions, unhindered by the smaller size of cancers currently ob­

tained.

Much has been written and even more stated orall y about the

lack of reproducibility of histologic grading. Its reliability

has rarely been scrutinized in a rigorous fashion as evidenced by

one widely quoted paper in which many of those involved were on

record as being opposed to grading, there was no learning set,

and no agreed-upon guidelines28 • With these three preconditions

to the study, its conclusions were pre-ordained. In other words,

with any system, if standards and definitions are not agreed

upon, then disagreement is certain.

10

The Carcinomas In-Situ

The two series of carcinomas in-situ, "ductal" and "lobular"

are distinctive in their histopathology and in their clinical

correlates . Before mammography, comedo DCIS was recognized as a

palpable mass which was usually present extensively within the

breast. Wi th screening mammography, irregular, linear microcal­

cifications herald s mall examples of DCIS which in many cases are

curable by the diagnostic biopsy, or by a second planned exci ­

sion. 35 These are followed by only rare recurrences which are

regularly found at the site of previous biopsy. Lagios et al

studied a large number of cases of DCIS, most less than 25 mm in

greatest diameter, detected by mammography and treated by

p l anned, wide excision without nodal dissection or radiotherapy.

The local recurrence rate of 10% was essentially confined to

those larger cases with advanced nuclear atypia and features of

comedo DCIS (Table 2). It is very clear from this experience

that the borderline between comedo and non-comedo is narrow, but

awaits a specific, consensus definition. With such an experience

of predicting clinical outcome, the definition should follow the

general outlines above and find wide acceptance. A careful sub-

classification of DCIS shows that more extensive higher grade

cases of DCIS, in the form of comedo and micropapillary types,

may show frequent widespread involvement and therefore these

types may not be amenable to local therapy 3 6 Solid and

cribri form types of DCIS are rarely multicentric or associated

11

with microinvasion. Similar to the careful reporting of special

types of invasive carcinoma, recognition of histologic types and

assessment of extent of DCIS yield extremely important prognostic

information.

Compared to DCIS, far less has been reported recently about

lobular carcinoma in-situ (LCIS) which has no distinctive mam­

mographic appearance but may be more common in breasts in which

microcalcifications are detected mammographically 37 • LCIS is

best thought of as a histologic marker of increased risk of sub-

sequent development of carcinoma38. This risk approaches ten

times that of comparable women in the general population, and is

most important for perimenopausal women. This risk is best ex­

pressed as absolute risk which is stated as 20 to 25% will

develop invasive cancer in about 15 years39,40. Whether or not

the risk remains stable beyond 15 years from the discovery biopsy

is unclear 41 but there is evidence that at least the relative

risk of cancer, after closely-related lesions, falls in the years

beyond 10- 15 years after discovery biopsy 42

Correlation with other factors of prognosis

Other methods than histol ogy have been proposed as predict­

ing outcome in breast carcinoma. These are varied and are under

intensive investigation . Receiving the greatest amount of atten­

tion in the literature are studies of various oncogenes, growth

fraction, and indicators of proliferative activity. Most of

12

these studies have not attempted to compare outcome indicated by

histology along with newer markers. An exception is a study by

Tsuda et al 43 in which the three components of the combined his-

tologic grade were compared to the presence of amplified copies

of c-erb B-2 gene. The copy number of c-erb B-2 was a prognostic

factor independent of tumor size and nodal status, and was -

strongly correlated with each component of the histologic grade.

Nodal status and grade we.re determinant of prognosis, with the

oncogene copy absorbed within the histologic grade after regres­

sion analysis43 • Both grade and amplification of c - erb B-2 were

strongly associ ated with aggressiveness of tumor rather than ex--tent of tumor spread. This result differs little from the paper

of Paik et al 44 who found that the presence of o-erb B-2

product identified about 4% of poor prognosis breast cancer cases

not identified by other means (particularly nuclear grade) .

Baak et al 45 found that morphometric features were stronger

prognosticators than c-erb B-2 protein overexpression. Similar

to Paik, overexpression of c-erb B- 2 identified a small group of

patients who died . Perren 46 has recently reviewed the prognostic

experience with c-erb B-2 in breast cancer and concludes that its

utility as a prognostic marker still needs to be tested relative

to others. certainly there is a strong correlation between his­

tologic features and c-erb B-2 expression as confirmed by Soomro

et al 47 who found that special histologic types of breast cancer

are regularly negative for c-erb B-2 .

13

Growth fraction and ploidy of c arcinomas may be determined

in several different ways. Dressler et al 48 were able to study

both DNA ploidy and s-phase f raction by flow cytometry using only

a small sample prepared from the same sample designated for

estrogen r eceptor analysis. By using a mathematical model,

S-phase fraction was determined in 81% of car cinomas analyzed . -

This is in contrast to previous methods in which S-phase fraction

could not be determined in aneuploid t umors. TUbiana et al were

able to show a strong correlation between tumor growth rate and

metastatic capacity 49. In this study, histologic grading was

also an independent prognostic indicator for relapse or death. A

s tudy -~t Guy 's Hospital 50 attempted to correlate flow cytometric

data with histologic grade. In a group of 140 primary breast

cancers, aneuploidy and high s-phase fraction were strongly as­

sociated with a high histologic grade. There was no significant

association between DNA ploidy or s -phase fraction and stage of

tumor (tumor s ize and nodal status) . Low proliferative activity

was associated with increased disease- free survival, overall sur-

vival, and survival after relapse . In contrast, there was no

significant association between ploidy and clinical course. A

multivariate analysis suggested that t he prognostic information

given by s-phase fraction, independent of tumor size and nodal

status, lost its independent significance when histologic grade

was included in the analysis50 • When t umor size, histologic

grade, estrogen receptor status, tumor ploidy, and s-phase f rac­

tion were examined in 169 patients with node-negative breast car-

14

cinemas, the prognostic factor which gave the wi dest separation

between subgroups was s-phase fraction 51 This was limited t o

tumors greater than 1 em in greatest diameter. By combining

tumor size and s-phase fraction, a significant difference was

found between those patients with tumors >1 em whose s-phase

fraction was <10% (5 year-relapse free survival of 78%) and those

with tumors >1 em with an s-phase fraction >10% (5 year RFS of

52%) 51 . We would conclude that ploidy is probably a weak in­

dicator of prognosis, while s -phase is a useful determinant of

proliferation intensity and prognosis.

A far less complicated method of estimating proliferative

activity of carcinomas is by use of monoclonal antibody Ki-67 and

immunohistochemical methods. This marker of cell proliferation

detects a nuclear antigen present in all phases of the cell cycle

except GO. The presence of immunostaining with Ki-67 correlates

with other measures of cell proliferation (S-phase fraction,

mitotic count) 52 Its presence was significantly higher in

aneuploid than in diploid carcinomas, and was associated with

high mitotic count and histologic grade. In addition, Ki-67

reactivity has been found to be inversely correlated with the

presence of estrogen and progesterone receptor 53 • Ki-67 seems to

be of independent prognostic value only if a high cutoff level is

selected 54 . This immunohistochemical method of detecting

proliferative activity is appealing for two reasons . First, no

tissue has to be separately analyzed apart from that submitted

for routine histologic examination. This is especially attrac-

15

tive in this age of smaller tumors . Related is the assurance

that the cells responsible for a positive reaction are indeed

within the carcinoma.

In summary, histologic parameters may yield vital prognostic

information as predictors of long term survival or short term

failure, proven in some settings. Features predicting long term

survival include the special types of carcinoma present in pure

form (Fig 1), especially tubular carcinoma. In those cases

which are of no special type, a long term survival may be ex­

pected for those patients with Grade I carcinomas which are lo­

calized to the breast and smaller that 20 mm; the survival is not

significantly different from age- matched female patients (91%

survival at 5 years, 88% at 8 years, 25 (Fig. 3). On the other

side of the spectrum (conversely), some indicators of short term

treatment failure are also well established. The Nottingham in­

dex identifies approximately 15% of patients who have a high

grade, high stage carcinoma associated with a poor survival (17%

at 5 years, 7% at 8 years) 25 The single feature of the com­

bined histologic grade weighing most heavily in determining

likelihood of death in the short term is high mitotic count

30,32, 33, a reflection of proliferative rate. A sensitive in­

dicator of proliferative activity, thymidine labeling index, has

been evaluated rarely in a setting allowing both long and short

term follow-up. TUbiana et al 49 support the strong indication

of a high proliferation rate for short term treatment failure,

16

but that at 8 years there was no difference between intermediate

and high proliferative rate groups. Even at 25 years, the low

proliferative group was still less likely to have metastases than

the other two groups. Histologic grade added further predictive

proven in this study.

The above studies indicate that valuable information may be

obtained methods other than histologic assessment alone.

However, all of these approaches require continuous evaluation

over a longer period of time and within different therapeutic

groups. None of these newer methods should, without concurrent

evaluations, supplant those assessments which have proven prog­

nostic worth. Without a clear understanding of the impact a par­

ticular parameter· may have on the clinical outcome, caution in

its interpretation is mandatory 55 • For example, an analysis of

ploidy among invasive carcinomas shows that the highest percent­

age (82%) of aneuploidy is present in cases with large areas of

necrosis, a reflection of rapid growth 56 on the other hand,

half of tubular carcinomas were reported to be aneuploid 56 • As

stated above 3 tubular carcinomas are an indication of an excel-

lent prognosis . This high incidence of ·aneuploidy in tubular

carcinoma may also be a reflection of differences in confines of

histologic definition, as the authors reported a preva·lence of

10.4%.

As we indicated earlier, some markers may be only meaningful

within certain stages. This is illustrated by a study of ploidy

of 8 encysted, noninvasive papillary carci nomas. Seven of these

17

were aneuploid and 1 was diploid 57 . Yet by well-def ined his­

tologic criteria, noninvasive papillary carcinoma is recognized

to be completely curable by local excision 58 . The aneuploidy,

if considered apart from the non- i nvasive status, could be con­

strued by some to be associated with a more malignant clinical

outcome.

Histol ogic preparations remain the mainstay for the diagnosis

of mammary carcinoma. They separate in situ from invasive le­

sions, and no other measure has even been suggested to take over

that important stratification. Some have used more subtle his-

tologic information (grading and special types) to give more

specif ic prognosti c information. This use of histologic data has

not found wide acceptance, in other words, the technology trans­

fer has been delayed. The reasons for this are probably many,

but analogy with other sites such as the prostate would indicate

that c linician demand is a necessi ty for pathologist acceptance.

Considering the ease of attaining the histologic preparations and

how easily c ases may be sent to numerous reviewers, it would seem

that we should be testing the utility of histologic preparations

to give us reliable prognostic information a.long with t .he more

recently developed candidates . This comes at a fortuitous time

i n the development of diagnostic histopathology, when there is a

dominant drive for the acceptance of rigorously defined specific

tissue patterns and cytologic features , and their linkage to

various cli nical outcomes.

18

>

TABLE 1

Distribution of Major Types of Infiltrating carcinoma*

Series NST** Lobular Medullary Tubular Mucinous Other

Rosen 59 75% 10% 10% 1% 2%

"Ductal "

Fisher, et 53% 5% 6% 1% 2% 28%

al60 "NOS"& Combined

Fu et 48% 11% 15% 7% 2% 5%

al61 "NOS" • Combined

Wallgren, 65% 14% 6% 7% 0

et a162

Sakamoto, 47% 2% 2% 0 4% 44%#

et a1 63

Edinburgh 70% 10% 5% 3% 2% 2%

series64 NST

*Series of invasive mammary carcinoma, modified from 64

**No special type

&Not otherwise specified

#Combined papillotubular and solid-tubular types

19

Combined

Legend for Table 1 :

Incidence and terminology of special types of invasive mam-

mary carcinoma among different investigators. Despite seeming

variations, there is major agreement on recognition of special

types. The difference may be related to differences in cohorts

as well as in confines of histologic definition.

•'

20

Table 2

Ri sk of Local Recurrence of DCIS after Bi opsy*

Subtype

Comedo

**

Cribriform

with necrosis

Cribriform

Nuclear Grade Necrosis

High +++

High +++

Intermediate +/-

Low 0

*Modified from Lagios, et al 65

Recurrence/Total

7/31

2/ 5

1/ 10

0/33

**This category was defined by Lagios as having some cribrifon

or papillary features wi th necrosis. However 1 the high nuclear

grade and necrosis place these few cases in a more broadl y

defined high-grade category , as is demonstrated by the 2/5 recur­

rence rate.

21

Figure Legend.s

Fig. 1: The 20-year survival figures of Dixon et a166 have

been extended 30 years. Special type carcinomas

have an excellent prognosis which remains stable over

the entire time course (reproduced with permission

from Dixon JM, Anderson TJ, and El ton, RA, University

of Edinburgh.

Fig . 2: Not inclu:led

Fig 3:

.•

Survival curves for patients based on Nottingham

prognostic index, which incorporates tumor grade, size

and nodal status. Reproduced with permission 25

22

..

g> ·;:; ·~ .80 :J

(/')

c: . . o 60

.. '[ . . e a. ~ .40

~ "3 § .20

·o

0.0

0 ·

S:Ulpson & Page

Special

Variant

NST

5 10 15 20 25 30

Time (Years)

FI GURE 1

S~son & Page

60 ~s

-40 ~ ·~ :l

(./)

eft.

20

----~6

0123456 7 8 Time (years}

FIGURE 3

References

1. Bartkova J, Barnes OM, Mill is RR, et al: Immunohistochemical

demonstration of c-erbB-2 protein in mammary ductal carcinoma in

situ. Hum Pathol 21:1164-1167, 1990

2 . Anderson TJ, Alexander F, Chetty U, et al : Comparative

pathology of prevalent and incident cancers detected by breast

screening. Lancet 1:519-522, 1986

3. Baker RR: Unusual lesions and their management. surg Clin N A

70:963-975, 1990

4 . Venable JG, Schwartz AM, Silverberg SG: Infiltrating

cribr iform carcinoma of the breast: A distinctive

clinicopathologic entity. Hum Pathol 21:333-338, 1990

5. Komari K, Sakamoto G, sugano H,H., et al: Mucinous carcinoua

of the breast in Japan: a prognostic analysis based on mor­

phologic features. cancer 61:989- 996, 1988

23

6. Rasmussen BB, Rose c, Christensen IB: Prognostic factors in

primary mucinous breast carcinoma. Am J Clin Pathol 87 : 155-160,

1987

7. Clayton F: Pure mucinous carcinomas of breast : Morphologic

features and prognostic correlates. Hum Pathol 17 : 34-38, 1986

8. Ridolfi RL, Rosen PP, Post A, et al: Medullary carcinoma of

the breast. A clinical pathological study with 10 year follow-up.

Cancer 40:1365-1385, 1977

9. Rapin V, Contesso G, Mouriesse H, et al: Medullary breast

carcinoma: a reevaluation of 95 cases of breast cancer with im­

flammatory stroma. Cancer 61:2503-2510, 1988

10. Wargotz E, Silverberg SG: Medullary carcinoma of the breast:

a clinicopathologic study with appraisal of current diagnostic

criteria. Hum Pathol 19:1340-1346, 1988

11. Fisher ER, Kenny JP, Sass R, et al: Medullary cancer of the

breast revisited. Breast Cancer Res Treat 16:215-229, 1990

12. Hilleren DJ, Andersson IT, Lindholm K, et al: Invasive

lobular carcinoma: Mammographic findings in a 10- year ex­

perience. Radiology 178:149-154, 1991

24

13. Dixon JM, Anderson TJ, Page DL, et al: Infiltrating lobular

carcinoma of the breast: An evaluation of the incidence and con­

sequence of bilateral disease. Br J surg 70 : 51 3 - 516, 1983

14. DiCostanzo D, Rosen PP, Gareen I, et al: Prognosis in in­

filtrating lobular carcinoma: An analysis of "classical" and

variant t umors. Am J Surg Pathol 14:12- 23, 1990

15. du Toit RS, Locker AP, Ellis IO, et al: Invasive lobular car­

cinoma of the breast- the prognosis of histopathological subtypes.

Br J cancer 60:605-609, 1989

16. Wheeler JE, Enterline HT: Lobular carcinoma of the breast in

situ and infiltrating. Pathol Annu llpt.2:161- 188, 1976

17. Schnitt SJ, Connolly JL, Recht A, et al: Influence of in­

filtrating lobular histology on local tumor control in breast

cancer patients treated with conservat ive surgery and

radiotherapy. Cancer 64:448-454, 1989

18. Stalsberg H, Thomas DB, Noonan EA, et al: Histol ogic types of

breast c ancer carcinoma in relation to international variation

and breast cancer risk factors. Int J Cancer 44:399-409, 1989

19. Henson DE: The histological grading of neoplasms. Arch Pathol

Lab Med 112:1091-1096 , 1988

25

20 . Henson DE: Cancer patient stag i ng: a perspective. Arch Pathol

Lab Med 115 : 120-121, 1991

21 . Azzopardi JG, Cbepick OF, Hartmann WH, et al: Histological

Typing of Breast Tumours (ed 2nd ). Geneva, World Health or­

ganization , 1981, pp 1-29

22. Elston cw, Ellis IO: Pathology and breast screening. His-

topathology 16:109-118 , 1990

-· 23. Elston CW: Grading of invasive carcinoma of the breast , in

Page OL, Anderson TJ . (eds): . Edinburgh, Churchill Livingston,

1988, pp 300-311

24. Page DL: Prognosis and Breast Cancer: Recognition of Lethal

and Favorab le prognostic types. Am J surg Pathol 15:334- 349, 1991

25. Todd JH, Dowle c , Williams MR, et al : Confi rmation of a prog-

nostic index in primary breast cancer. Br J Cancer 56:489-492,

1987

26. Black MM , Speer FD : Nuclear s t ructure in cancer tissues. Surg

Gynecol Obstet 105:97-102, 1987

26

27. Fisher ER, Redmond c, Fisher B, et al: Pathologic findings

from the National Surgical Adjuvant Breast and Bowel Projects

(NSABP): Prognostic discriminants for a-year survival for node­

negative invasive breast cancer patients. Cancer 65 Suppl.:2121-

2128, 1990

28. Gilchrist KW, Kalish L, Gould VE, et al: Interobserver

reproductibilty of histopathological features in stage II breast

cancer. Breast Cancer Res Treat 5:3, 1985

29. McGuire WL, Tandon AK, Allred DC, et al: How to use prognos­

tic factors in axillary node-negative breast cancer patients.

JNCI 82:1006-1015 , 1990

30. Parl FF, Dupont WD: A retrospective cohort study of his­

tologic risk factors in breast cancer patients. Cancer 50:2410-

2416, 1982

31. Baak JPA, Van Dop H, Kurver PHJ, et al : The value of mor­

phometry to class i c prognosticators in breast cancer. Cancer

56 : 374-382, 1985

32 . Joensuu H, Toikkanen S, Klemi PJ : DNA index and S-phase frac­

tion and their combination as prognostic factors in operable duc­

tal breast carcinoma. cancer 66:331-340, 1990

27

33. Baak JPA: Mitosis counting in tumors. Hum Pathol 21:683- 685,

1990

34. Quinn CM, Wright NA: The clinical assessment of proliferation

and growth in human tumours: Evaluation of methods and applica­

tions as prognostic variables. J Patho1 160:93- 102, 1990

35. Lagios MD, Margolin FR, Westdahl PR, et al: Mammographically

detected duct carcinoma in situ. Cancer 63 :618-624, 1989

36. Patchefsky AS, Schwartz GF, Finke l stein SO, et al:

Heterogeneity of intraductal carcinoma of t he breast. Cance r

64:731- 741, 1989

37. Pope TL, Fechner RE, Wilhelm MC, et al: Lobular carcinoma in

situ of the breast:Mammographic features. Radiology 168 : 63-66 ,

1988

38 . Gump FE: Lobular carcinoma in situ: Pathology and treat.ment.

Surg Clin N A 70 : 873-883, 1990

39. Page DL, Dupont WD: Premalignant conditions and markers of

elevated risk in the breast and their management. Surg Clin N A

70 :831-851 , 1990

28

40. Shack RB, Page DL: The patient at risk for breast cancer:

pathologic and surgical considerations. Perspectives in Plast

surg 2:43-62, 1988

41. Page DL, Kidd TE, Dupont WD, et al: Lobular neoplasia of the

breast : Higher risk for subsequent invasive cancer predicted by

more extensive disease. Hum Pathol 1991(in press)

42 . Dupont WD, Page DL: Relative risk of breast cancer varies

with time since diagnosis of atypical hyperplasia. Hum Pathol

20:723-725, 1989

43. Tsuda H, Hirohashi s, Shimosato Y, et al: correlation between

histologic grade of malignancy and copy number of c-erbB-2 gene

in breast carcinoma: A retrospective analysis of 176 oases. Can­

cer 65:1794-1800, 1990

44 . Paik s, Hazan R, Fisher ER, et al: Pathologic findings from

the National Surgical Adjuvant Breast and Bowel Project: Prog­

nostic signifi cance of erbB- 2 protein overexpression in primary

breast cancer. J Clin Oncol 8 : 103-112, 1990

45. Baak JPA, Chin D, van Diest PJ, et al: Comparative long-term

prognostic value of quantitative HER- 2/neu protein expression,

DNA ploidy, and morphometric and clinical features in paraffin­

embedded invasive breast cancer. Lab Invest 64:215-223, 1991

29

46. Perren TJ: c-erbB-2 oncogene as a prognostic marker in breast

cancer. Br J cancer 63:328-332, 1991

47. Soomro s, Shousha s, Taylor P, et al: c-erbB-2 expression in

different histological types of invasive breast carci noma. J Clin

Pathol 44:211-214, 1991

48. Dressler Lynn G, Seamer Larry c, Owens Marilyn A, et al: DNA

flow cytometry and prognostic factors in 1331 frozen breast can­

cer specimens. cancer 61:420-427, 1988

49. Tubiana M, Pejovic MH, Koscielny s, et al: Growth rate,

kinetics of tumor cel l proliferation and long- term outcome in

human breast cancer. Int J Cancer 44:17-22, 1989

50. O'Reilly SM, Camplejohn RS, Millis RR, et al: Proliferative

activity, histological grade and benefit from adjuvant

chemotherapy in node positive breast cancer. Eur J cancer

26:1035-1038, 1990

51. O'Reilly SM, camplejohn RS, Barnes OM, et al: Node-negative

breast cancer: Prognostic subgroups defined by tumor size and

f l ow cytometry. J Clin Oncol 8:2040-2046, 1990

30

52 . Isola JJ, Helin HJ, Helle MJ, et al: Evaluation of cell

proliferation in breast carcinoma: Comparison of Ki-67

munoh i stochemical study, DNA flow cytometric analysis,

mitotic count. cancer 65:1180-1184, 1990

im­

and

53 . Di Stefano 0, Mingazzini PL, scucchi L, et al: A comparative

study of histopathology, hormone receptors, peanut lectin bind·

ing, Ki -67 i mmunostaining, and nucleolar organizer region­

associ ated proteins in human breast cancer. Cancer 67 : 463-471,

1991

54 . Wintzer H-0, Zipfel I, Schulte-Mnting J, et al : Ki-67 im­

munostaining in human breast tumors and its relationship to prog­

nosis. Cancer 67:421-428, 1991

55 . Levi ne MN, Browman GP, Gent M, et a l : When i s a prognostic

factor useful?:a guide for the perplexed. J Clin Oncol 9:348-356,

1991

56 . Pontn J, Holmberg L, Trichopoulos o, et al : Chapter I :

Biology and natural history of breast cancer. Int J Cancer 46

Suppl . 5:5-21, 1990

57. Corkill ME, Sneige N, Fanning T, et al: Fine-needle aspira­

tion cytology and . flow cytometry of intracystic papillary car­

cinoma of breast. Am J Clin Pathol 94:673-680, 1990

31

58. carter 0, Orr SL, Merino MJ: Intracystic papillary carcinoma

of the breast. After mastectomy,radiotherapy or excisional biopsy

alone . Cancer 52:14-19, 1983

59 . Rosen PP: The pathological classification of human mammary

carcinoma: past,present and future . Ann Clin Lab Sci 9:144-156,

1979

60. Fisher ER, Gregorio RM, Fisher B: The pathology of invasive

breast cancer. A syllabus derived from findings of the national

surgical adjuvant breast project (Protocol no 4). Cancer 36 : 1-85,

1975

61. Fu YS, Maksem JA, Hubay CA, et al: The relationship of breast

cancer morphology and estrogen receptor protein status, in

Fenoglio CM, Wolff M. (eds}: Progress In Surgical Pathology. New

York, Masson, 1981, pp 65-76

62. Wallgren A, Silfverward C, Eklund G: Prognostic factors in

mammary carcinoma. Acta Radiol 15:1-16, 1976

63 . Sakamoto G, Sugano H, Hartmann WH: Comparative pathological

study of breast carcinoma among American and Japanese women, in

McGuire WL. (ed}: Breast Cancer . Advances in research and treat­

ment . (ed 4th). New York, Plenum, 1981, pp 211-231

32

64. Page DL, Anderson TJ, Sakamoto G: Infiltrating carcinoma:

major histological types, in Page DL, Anderson TJ. (eds}: Diag­

nostic Hi stology of the Breast. Edinburgh, Churchill Livingstone,

1987, pp 193-235

65. Lagios MD: Duct carcinoma in Situ: Pathology and treatment.

Surg Clin N A 70:853-871, 1990

66. Dixon JM, Page DL, Anderson TJ, et al: Long-term survivors

after breast cancer. Br J surg 72:445-448, 1985

33

16

by David L. Page, MD, and William D. Dupont, PhD, Nashville, TN

In 1986, David Winchester, MD,. FACS, published in the Bulletin the consensus statement of the College of American Pa­thologists, supported by the American

Cancer Society, on the relative risk of invasive breast carcinoma based on histopathologic ex­amination of benign breast tissue.• He noted that surgeons are confronted daily with women concerned about lumps or lumpy areas within the breast and, of course, confronted with di­agnoses from mammograms tlult cannot give a certain diagnosis of benign or malignant. Somewhere within this spectrum of physical examination and mammographic examination there rests a borderline of histopathology as well. This borderline of histopathologic defi­nition may now provide indications of in­creased cancer risk. Specific histopathologic lesions indicate varied levels of subsequent risk of the development of invasive breast cancer (Table I on page 181, and have in the last five years gained further acceptance and corrobor-ation from other analogous studies. ,

The original intent of the 1985 consensus conference in New York was to defuse or re­duce a widely held belief that the clinical de­tection of "fibrocystic disease" indicated some meaningful link to breast cancer.2 The consen-

VOI..\JME 76. N'Ulii(B£R t . Al(£1UCA.N COIJ..tCE OP SUROEONS BUI.l..ETIN

sus conference noted that some histologic di­agnoses had links to greater relative cancer risks, but that most chaoges did not elevate risk, including cysts and fibrosis. In fact, in the era before mammography, 70 Jlercent of biopsied women did not differ in cancer risk from the general population, and another 20-25 percent had an elevated risk in the 15 years after biopsy, which only approached a dou­blin.g of the reference population used.

One major change since 1985 ha:s been in the placement of well-developed examples of sclerosing adenosis within the slightly in­creasedriskcategwy."-4 There may still be some question about where this change actually be­longs, but as this article will indicate, the slightly increased risk category has no great clinical importance. In any case, these low-risk patients are not women who should be treated very differently from the general population.

Somewhat fewer than 5 percent of women in a premammographic series bad specific pat­terns of atypical hyperplasia (AH) that ap­proached the pa~ of carcinoma in situ <CISJ."'' These women with AH had a risk of cancer four to five times greater thao that of the general population, or about one-half the risk associated with microscopic carcinoma in

situ. When general criteria from original diag­noses of many pathologists were used, AH still elevated risk, but only to about three times. 7

Several recent studies support the utility and predictiveness of specifically defined AH. •.s.• Fol­low-up studies of comparable type involving women with mammographically detected lesions are not available. However, it is clear that the incidence of AH is higher in mammographically directed biopsies.10 The primary therapeutic im­plications of these premalignant lesions are in­tensified breast cancer surveillance and screening for these patients.

Further studies have indicated an appreciable interaction between atypical hyperplasia and other nonanatomic risk factors, particularly a family history of breast cancer. 7 Also, lower dos­age estrogen replacement (specifically conju­gated preparation) after menopause does not appear to further affect risk in any histologically defined group. 11

Implications of increased clinical risk Patients are only considered at elevated risk

for the development of carcinoma if the risk is reliably determined to approach double that of the general population. Lesser degrees of risk may be important in nonclinical settings, but are not considered here. Even a doubling of the risk may not be of direct clinical relevance because the practical impact of the risk may be viewed dif­ferently within the circumscribed framework of the ensuing few years (Table II on page 19).

Elevated risk of breast cancer is of unques­tioned clinical importance when its magnitude approaches five times the general population. Comments on magnitude of relative risks are in­herently confusing without an immediate refer­ence group. This reference group is usually the general population of women without the risk factor. The age of a patient and the number of years at risk are important in evaluating the im­pact that these risk statements should have in the clinical setting. Thus, in the clinical setting of a single patient, absolute risk statements are more relevant. However, these latter statements must be known to be relevant to similar women over a similar period of time.

The relationship between relative risk and ab­solute risk is poorly understood by many people.

The following question is not uncommon: If an average woman's lifetime risk of breast cancer in 'North America is 10 percent, and if my patient is diagnosed as having a breast cancer relative risk of five, does this mean that my patient has a 50 percent lifetime risk of breast cancer?

The answer to tliis question is no, but the ex· planation requires an understanding of what we mean by relative risk, and how probabilities are combined to give lifetime absolute risk. By reJ. ative risk, we mean a risk ratio between two groups of patients over a short period of time. The risk that a 42-year-old woman will develop breast cancer in the next year is about one in a thousand. If your 42-year-old patient has a five­fold increase in breast cancer risk compared to the general population, then her chances of de­veloping breast cancer in the next year are five in a thousand. Relative risks often vary with time. Let's suppose that your patient's relative risk will remain constant at five for the remainder of her lifetime. What is her probability of eventually developing breast cancer? This is simply the sum of the probabilities of her first developing breast cancer in each ofthe remaining years of her life. Consider her probability of first developing breast cancer in any one of these years-say her 7lst year. To first develop breast cancer at age 70, there are two criteria:

1. She must survive to her 70th birthday with­out developing breast cancer.

2. She must then develop breast cancer in her 7lst year.

The probability of number 2 is five times the breast cancer risk of a 70-year-old from the gen. eral population. The probability that she firSt de­velops breast cancer at age 70 equals the probability of number l times the probability of number 2.

If the probability of number l were equal to her probability of surviving until her 70th birth· day, then her lifetime risk would indeed be five times the risk that a 42-year-old from the gen· era! population will eventually develop this dis· ease. In fact, breast cancer is sufficiently common that her chances of developing breast cancer be­fore age 70 are appredable. This fact reduces the probability of number 1, which in turn reduces her lifetime risk of breast cancer. The probability of number l for rare diseases is closely approxi· t

S&Fr£MBER 1991 ACS 8 U1.LE'Tri

No increased risk; Women with any lesion specified below in a biopsy specimen are at no greater risk for invasive breast carcinoma than comparable women who have had no breast biopsy:

Adenosis Duc:t ectasia Apocrine metaplasia Fibroadenoma Cysts, macro and/or micro Fibrosis Hyperplasia (mild, more than Mastitis (inflammation)

2 but not more than 4 Periductal mastitis epithelial cells in depth) Squamous metaplasia

Slightly increased risk (1.5 to 2 times): Women with any lesion specified below in a biopsy specimen are at slightly increased risk for invasive breast carcinoma relative to comparable women who have had no breast biopsy:

Hyperplasia, moderate or florid, solid or papillary Papilloma with fibrovascular core Sclerosing adenosis, well-developed

Moderately increased risk ( 4 to 5 times): Women with a lesion specified below in a biopsy specimen are at moderately increased risk for invasive breast carcinoma relative to comparable women who have had no breast biopsy:

Notes

Atypical hyperplasia (borderline lesion) Specific patterns of atypical ductal hyperplasia Specific patterns of atypical lobular hyperplasia

A. All forms of adenosis were accepted in 1985 as having no indication of increased cancer risk. Since that time, well-developed examples of sclerccing adenosis, apart from other associations, indicate a slightly increased risk.. 3 ·•

8. Cysts were placed in this category in 1985 and probably should remain there. although there is some interest" in analyzing special subsets of cysts identified biochemically or by apocrine cytology. These studies ue in progress. Some physicians suggest that women with a family bistory and cysts (presumably large Md palpable) have a slightly greater risk than identified by their family history alone, but the effect does not as much as double the familial indicator, and bas not been controlled for the simultaneous presence of hyperplastic lesions.

C. Some large epidemiologic studies suggest that women with fibroadenomas (FA) have a slightly in· creased risk of later carcinoma. about 1. 7 times. This is "slight," less than double that of comparable women. Note that the absolute risk of cancer for these young women is not greatly changed, i.e., women

18 under 40 years old in NGrth America have less than a 0.1 percent incidence of breast cancer per year.

VOLUME 76. NUMBER 9, AMERICAN COLLEGE Of SVROEONS JliJl.LtttN

mated by the patient's absolute probability of survival to the given age. Thus, for rare diseases we can multiply relative risk by lifetime risk if we are willing to assume that relative- risk re­mains constant throughout life. As a disease be­comes more common, this approximation becomes worse, and becomes absurd for sufficiently com­mon outcomes.12 For example, the lifetime risk of cardiovascular death is in excess of 50 percent, but smokers with a tenfold relative risk of this fate do not have a 500 percent absolute risk of cardiovascular death. Indeed, their risk must be less than 100 percent since some smokers will die of other causes.

"Increased cancer risk" is assigned to those conditions that have a statistically significant el­evation in cancer risk of a magnitude that ap­proaches twice that of the general population controlled for age and number of years at risk. It is important not to extrapolate such risks to time intervals that are much longer than those doc­umented by follow-up studies in the literature. The preceding discussion, in particular, indicates that projecting risk over the entire lifetime of a patient is misleading and usually overstates the magnitude of the patient's absolute risk. 12.13

All individually and clinically relevant risk statement is made in terms of absolute risk (AR). This states the probability of an event occurring in a certain period of time. Absolute risk for AH is stated as: 10 percent in 10-15 years. In general, we do not believe that breast cancer risk should be atended beyond that length of time because the stability of risk with time is unproven.13 It is our experience, particularly with older women, that these elevated risks will fall 10 to 15 years after detection. 13 Thus, AR should be used pref­erentially in the clinical setting with an individ­ual patient, but should be carefully sculpted to the setting. AR may be derived from relative risk (RR).12 However, overestimates of AR will occur even if the relative risk is assumed to be constant for a full20 years, because, in fact, RR often di­minishes with time after biopsy. This is precisely what happens for proliferative disease. 13 Thus, it is the AR that is most useful in the clinical set­ting, but it must be derived with care for each patient, assessing the relevance of published fig. ures to the specific patient.,.

The diagnostic phrase proliferatiue breast dis-

40 50

Table n ·

Absolute risk" of invasive. breast cancer ·. ·

in 12 months

Usual (RRb = 1)

111,000 2.2/1,000

Slight increase fBR = 2l

2/1,000 4.4/1,000

a. For most women in North America b. Relative risk

ease (PBDJ indicates that there are proliferative alteratioll8 noted by histology, and that they io­dicate a disease by their demonstrated link to an increased risk of subsequent carcinoma deveJ. opment.16 PBD is a phrase that ma,y be used to knit together the histologic and risk statements found in Table I for slight and greater risk le­sions. The greatest use of tbe diagnostic phrsse is in the clinical "negative for PBD'' or "no PBD.• These phrases indicate that lesioll8 with cancer risk implications have been sought, but not found.

Slightly i~ rWl The magnitude of risk elevation in this slightly

elevated group is reliably increased more than 50 percent over that of women of similar age from the general populition. However, risk of deveJ. oping invasive carcinoma does not reliably attain the magnitude of 100 percent greater or double ~t of the reference population. This nmge of risk may be recorded as 1.5-2.0 times that of the general population, and emphasizes that risk as­sessment or assignment is not precise, but is rather probabilistic and should be understood to include a range of values. The magnitude of the relative risk depends largely on the populations used for comparison. Thus, we found a risk ele­vation of 1.6 times when the general population was used as a comparison group, and 1.9 when women from our study with only mild or no hy· perplasia were the comparison population. 5

The ~or histologic patterns and categories

contained in this slight elevation of risk categOry are the more developed, usual, or common types of epithelial hyperplasia, which may be termed PBD without atypia (PDWAl. The terms papil­lomatosis and epithelicsis have been used for ihese patterns of hyperplasia (Table Il. 18 These latter terms will still find use, but have caused confu· sion and are inconsistently applied in different countries. The term usual relays the idea that these are the commonly found patterns of cytol­ogy and cell relationships seen when cell num­bers are increased within the basement membrane-bound spaces within the human breast. These alterations are most common in the im­mediate premenopausal ages. The 1985 consen­sus de.fined this histologic pattern as having a more extensive degree of epithelial proliferation than four cells above a basement membrane. This is largely true, and serves as a good rule of thumb in separating mild hyperplasia (epithelial lining of two to four cells deep with no increased cancer risk) from lesions with slight cancer risk impli­cations.

AlsoTesiding within the slightly increased cat­egory are the histologic diagnoses of sclerosing adenosis and papilloma. Reticence for acceptance of sclerosing adenosis by the consensus confer­ence in 1985 rested on the fact that it was not repeatedly demonstrated as an indicatOr of in­creased risk, as well as that, while it was in­cluded in the group of PDWA lesions initially, careful analyses of associated and confounding elements had not been performed.5 Since that time we have completed an analysis of sclerosing ad-

Dr. Page is professor of pat/wlogy and director of anctomic pat/wlogy, Varukrbilt Urm'f!rsity Medical Center, Nashville, TN.

enosis and its various associations. When rigor­ously deftned histologically, sclerosing adenosis is an indicator of slightly increased risk of sub­sequent breast carcinoma apart from its associ· ation with other risk indicators. • McDivitt and associates have also found sclerosing adenosis to be an indicatOr of increased risk -double that of the general population. 3

The solitary papilloma has been thought to have little concern relative to subsequent breast can­cer risk, and this is probably consistent with its placement within the slight risk category. Carter suggested that follow-up of women after removal of papillomas indicated a somewhat increased riskP Our studies have conftrmed this link of papillomas with increased risk of slight magni­tude.5

Moderately increased mk This term was originally chosen by the con­

sensus conference to place these lesions in per­spective between those noted previously and microscopic examples of in situ carcinoma. The relative risk for subsequent invasive carcinoma of the atypical hyperplasias within this group is four to five times tha,.t of the general population. This is approximatefy half the risk experienced by women with microscopic in situ carcinoma.

The recognition that a woman has a fourfold increase in cancer risk relative to the general population demands clari.fication because it is easily misunderstood. Such relative risk figures compare two groups for purposes of statistical evaluation, and are not readily transferred di­rectly to clinical practice. Any relative risk fig­ure is bound by the experience of the groups studied. Since few studies of benign breast dis­ease have had more than 15 years of follow-up, these statistics only apply to 10 to 15 years after biopsy. It is also somewhat confined to the group of women who are most frequently biopsied in usual clinical practice-that is, those women about the age of 50. Few women of the ages less than 30 and more than 60 were present in our studies, and risk ftgures for atypical lesions must be understood to be less certain for women in these younger and older age groups. 14

The atypical hyperplastic lesions that com­prise this moderate risk group are deftned by their histologic resemblance to lesions long recognized

as carcinoma in ;\tu. They are named by analogy to lobular carcinoma in situ and ductal carci­noma in situ, respectively. The atypical hyper­plasias, as defmeil in this manner, may be viewei:i as having some of the same features as the car­cinoma in situ lesions. but in less t han fullv de-veloped form. 6 ·18 • -

The suggestion followed by most practitioners at this time is that women in the slightly -ele­vated risk group should merely be encouraged to have a yearly mammogram and annual physical examination after t he age of (approximately) 40. Self-examination should also be encouraged in these women. As these· same recommendations are given to women with no known level of risk, women with slightly increased risk should not be encouraged to be overly anxious.

There is such a strong interaction with family history and AH that it is Televant to consider women with atypical hyperplasia who have a positive family history iFH) of bre~t cancer sep­arately from those who do not. The definition of a positive FH in these studies was at least a first degree relative \mother, daughter, sister) with proven breast cancer. The absolute risk of breast cancer development in women with atypical hy­perplasia without a family history was 8 percent in 10 years, whereas those with a positive family history experienced a risk of about 20-25 percent at 15 years. This strong interaction with family history has been supported in a recent study.' This magnitude of risk is closely analogous to that accorded lobular .carcinoma in situ. 19•22

. Probably the most important single determin­ing factor for clinical management is the worn· an's level of concern. It is overly dramatic to call this variable ''cancerphobia," but that is the most familiar tenn. Our fervent belief is that one should at least follow the precept that, once a moderate or high-risk lesion is determined, no decision should be rushed. It is not a medical emergency, or an emergent situation of any type. Even the carcinomas in situ of well-developed magnitude and extent have not been demonstrated to threaten life within several months following bi· opsy. Indeed, the "latent" period between detec· tion of the microscopic ductal carcinomas in situ, as well as lobular carcinoma in situ biopsy, not treated further and the a ppearance of invasive carcinoma averages almost 10 years or more. 20·23.24

Thus, no harm is likely to result if several months pass between the detection of a high or moder· ately high-risk lesion and the formalization of a clinical decision. Such decisions are best made without the overlay .of immediate drama and stress. They should also be made with full knowl­edge of what the impact of. this diagnosis and course of action will have upon the woman's quality of life.

However difficult it might be for one to under­stand risk elevation in the sense of a group or defined population, it is certainly even more dif· ficult to do so in the case of an individual. Thus, if the individual's level of conce.rn is extremely high {such as impairing sleep), the patient and her physicians v.ill more seriously consider some type of mOdified prophylactic mastectomy on the indication ofreducing risk.25

Mammographic surveillance is widely ac­cepted as a clinical alternative over extirpative surgery for most women. The logic of this deci, sion rests largely in the knowledge that: (1') mas· tectomy would r emove many more breasts than ever would develop cancer, (2) the patient's breast cancer risk will more closely approximate that of sligbtly~elevated risk if she remains free of can· cer for 10 years after biopsy, indicating moderate risk, 13 and (3) the current era of mammography should improve the good posttreatment prognosis for developing breast cancer in women with mod­erate and high risk followed closely by palpa· tion.26

There is no generally accepted program for clinical surveillance of women with moderate risk.

Dr. Dupont is Cl$SOCiate professor of . pret•entive medicine

and director, division of biostatistics.

Vanderbilt University Medical Center, Nashville. TN.

SEPT£~i8ER J99J ACS BULL.ET!N

21

However, yearly mammography along with clin­ical examination once or twice yearly should be considered prudent along with a recommenda­tion for breast self-examination"' Whether more frequent mammographic examinations would be practical is questionable. Some surgeons have suggested adding a physical examination at the midpoint between yearly examinations to rein· force vigilance. A repeat mammogram at five-to­six months after initial biopsy is useful because the possible effect of scarring at the-biopsy site may be assessed and the early evaluation of a possible coexisting carcinoma may be accom­plished. This suggestion is analogous to the guidelines. following radiation therapy. 25

It is abundantly clear that many factors will impinge upon the clinical management of the woman with high risk of AH and family history, Although indications for biopsy (surgical or needle) will remain the same as for other women, it is also clear that biopsies will be performed more frequently' in this group for less certain in­dications. If frequent biopsies are performed, then preventive surgical strategies such as prophylac· tic mastectomy should be more seriously consid­ered,25 and will certainly be more frequently considered in women with a 20 percent risk of invasive carcinoma in the next 15 years, espe· cially if they have had the emotional trauma of witnessing their mothers die. In any case, even bilateral mastectomy need not be considered emergently. A six-month delay in decision-mak­ing is not likely to put life at any risk.

Conclusi<>n8 Clinical assessment of breast cancer risk is an

endeavor unfamiliar to most physicians. Our fa. miliarity with risk. assessment will grow in the ensuing years. Of greatest importance to this growth will be the garnering of more specific in' formation about determinants of risk and their interaction with screening and therapeutic mo­dalities. The latter will involve intervention trials for prevention of breast cancer.

At the present, women with slightly increased risk should be encouraged to follow a regular !yearly) program of mammographic surveillance, while women with. lesions associated with mod­erately increased risk should follow such a pro·

22 grani without fail. Atypical hyperplasia is rare,

VOl.\.~(£ 76. NUMBER 9, A.'d.£RIC.4.."i COLLEGE OF SURCEOSS 8 1J"l.l.&T1N

occurring in only 4 percent of breast biopsy spec· imens prior to the mammographic era, although this incidence is currently higher in mammo­graphically directed biopsies."• Other consider­ations such a:s mammographic density, family history, and high anxiety will impinge on clinical man.agement decisions. In most of these patients, however, extirpative surgery is probably not a frequent practical consideration. 25 !Iii

This work was supported by National Cancer Institute contract N01-CB-74098 ancigrants. ROl C.-'.-28223. R01 CA-31698. and ACS RD 222.

References

1. Winchester DP: Conse~us statement. The rela­tionship oi fibrocystic disease to breast cancer. Buil Am C<Jil Surg, 71:29-31, 1986.

2. Hutter "RVP: Conse~us meeting. Is "fibrocystic &sease" of the breast precancerous? Arch Pathol Lab Med, 110:171-173, 1986.

3. McDivitt, RW, Rubin GL, Stevens JA. Wingo PA: Benign breast disease histology and the risk of breast cancer. Lab /nuest, 58:62A, 1988.

4. Jensen RA. Page DL. Dupont WD. Rogers LW: Invasive breast cancer risk in women with scle­rosing adenosis. Cancer. 64:1977-1983. 1989.

5. Dupont WD, Page DL: Risk factors ior breast can­cer in women with proliferative breast diseast!. New Eng J Med, 312:146-151. 1985.

6. Page DL, Dupont WD, Rogers LW. Ra.dos MS: Acypical hyperplastic lesions of the female breast. A long-term follow-up study. Cancer, 55:2698-2708. 1985.

7. Carter CL, Corle OK. Micozzi MS: A prospective study of the development of breast cancer in 16,692 women with benign breast disease. Am J Epide­miol, 128:467-477, 1988.

8. Tavassoli FA, Norris. HJ: A comparison of the. re­sults of long-term follow-up for atypical intra­ductal hyperplasia and intraductal hyperplasia of the breast. Cancer, 65:518-529. 1990.

9. Connolly JL, SChnitt SJ, London SJ. Colditz G: Berugn breast disease and risk of subsequent breast canoor: 1'he experience in the nurses' h.alth study. Lab Inuest, 64:10a, 1991 (abstracu.

10. Rubin E, Alexander RW. Visscher OW. et al: Pro­liferative disease and atypia in biopsies performed for mammographically detected nonpalpable le­sions. Cllflcer, 61:2077-2082. 1988.

11. Dupont WD. Page DL. Rogers LW, Pari FF: Influ­ence of exogenous estrogens, proliferative breast disease. and other variables on breast cancer risk. Can«r. 63:948-957.1989.

12. Dupont WD: Converting relative risk.s.to absolute risks: A graphical approach. Stat Mtd. 8:641-65!. 1989.

13. Dupont WD. Page OL: Relative risk of breast can· cer varies with time since diagnosis of atypical hyperplasia. Hum Pathol, 20:723-725, 1989.

14. Page DL. Dupont WD: Histopathologic risk factors for breast cancer in women with benign breast disease. Sem Surg One, 4:213-217, 1988.

15. Page OL. Dupont W'D: Anatomic markers of hu­man premalignancy and risk of breast cancer. Cancer, 66:1326·1335, 1990.

16. Azzopardi. J : Problems in. bretUI pathology. Phil­adelphia. P A: Saunders. 1979.

17. Carter 0: Intraductal papillary tumors of the breast: A study of 78 cases. Cancer, 39: 1689· 1692. 1977.

18. Page DL. Anderson TJ. Rogers LW: Epithelial hy­perplasia. In: Page DL, Anderson TJ (eds.J: Di­agnostic histopathology of the brt<Ut. Edinburgh: Churchill LivingstOne, 1988.

19. Wapnir IL. Rabino-..-itz B. Greco RS: A reappraisal of prophylactic mas<ectOmy. Surg Oynecoi & Db­stet. 171:171-184. 1990.

20. Frykberg ER. Santiago F. BetsiU WL Jr .. O'Brien PH: Lobular carcinoma in situ of the breast. Surg Oynecol & Obstet. 164:285-301, 1987.

21. Gump FE: Lobular carcinoma in s itu: Pathology and treatment. Surg Clin N Am, 70:873-883. 1990.

22. Hutter RVP: The management of patients v.;th lobular carcinoma in situ of the breast. Cancer, 53:798-802. 1984.

23. Betsill WL Jr .. Rosen PP. Li"berman PH. Robbins GF: Intraductal carcinoma. Long-term follow-up alier treatment by biopsy alone. JAMA. 239:1863-1867, 1978.

24. Page DL. Dupont WD. Rogers LW, Landenberger M: Intraductal carcinoma of the breast: Follow-up alier biopsy only. Cancer. 49:751-758, 1982.

25. Shack RB, Page DL: The patient at risk for breast cancer: Pathologic and surgical considerations. Per: "' Pla8t Surg, 2:43-62. 1988.

26. Ham: nsen CD, Lane N, Latle$ R, Bodian C: Lob-

ula.r neoplasia !so-called lobular carcinoma in situi of the breast. Cancer, 42:737-769. 1978.

27. Locker Al', Caseldine J. Mitchell AK. et al: Re· suits from a seven-year programme of breast self­examination in 89.010 women. BrJ Cancer, 60:401-405, 1989.

28. Mendelson EB: imaging the post-surgical brea.<t. &m. Ultra$ CT MR, 10:19-35, 1989.

'29. Ande1:10n TJ: Premalignant benign disease-Fact or fiction? Br J Clin Pract, 56:31 !51-35 (5 ). 1988.

..

SE:PTDr8P I ttl ACS 8ULL.£ns

23