Download - Nej Mo a 1000067
-
7/30/2019 Nej Mo a 1000067
1/13
original article
T he n e w e n g l a n d j o u r n a l o f medicine
n engl j med 363;7 nejm.org august 12, 2010640
Reduced Treatment Intensity in Patients
with Early-Stage Hodgkins LymphomaAndreas Engert, M.D., Annette Pltschow, Ph.D., Hans Theodor Eich, M.D.,
Andreas Lohri, M.D., Bernd Drken, M.D., Peter Borchmann, M.D.,Bernhard Berger, M.D., Richard Greil, M.D., Kay C. Willborn, M.D.,Martin Wilhelm, M.D., Jrgen Debus, M.D., Michael J. Eble, M.D.,
Martin Skler, M.D., Antony Ho, M.D., Andreas Rank, M.D.,Arnold Ganser, M.D., Lorenz Trmper, M.D., Carsten Bokemeyer, M.D.,
Hartmut Kirchner, M.D., Jrg Schubert, M.D., Zdenek Krl, M.D.,Michael Fuchs, M.D., Hans-Konrad Mller-Hermelink, M.D.,
Rolf-Peter Mller, M.D., and Volker Diehl, M.D.*
The authors affiliations are listed in theappendix. Address reprint requests toDr. Engert at Kerpenerstr. 62, 50937 Kln,Germany, or [email protected].
*Additional investigators in the HD10study are listed in the SupplementaryAppendix, available with the full text ofthis article at NEJM.org.
N Engl J Med 2010;363:640-52.Copyright 2010 Massachusetts Medical Society.
A B S T RA C T
BACKGROUND
Whether it is possible to reduce the intensity of treatment in early (stage I or II)Hodgkins lymphoma with a favorable prognosis remains unclear. We therefore con-ducted a multicenter, randomized trial comparing four treatment groups consistingof a combination chemotherapy regimen of two different intensities followed byinvolved-field radiation therapy at two different dose levels.
METHODS
We randomly assigned 1370 patients with newly diagnosed early-stage Hodgkins lym-phoma with a favorable prognosis to one of four treatment groups: four cycles ofdoxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) followed by 30 Gy of
radiation therapy (group 1), four cycles of ABVD followed by 20 Gy of radiationtherapy (group 2), two cycles of ABVD followed by 30 Gy of radiation therapy (group 3),or two cycles of ABVD followed by 20 Gy of radiation therapy (group 4). The primaryend point was freedom from treatment failure; secondary end points included effi-cacy and toxicity of treatment.
RESULTS
The two chemotherapy regimens did not differ significantly with respect to freedomfrom treatment failure (P = 0.39) or overall survival (P = 0.61). At 5 years, the rates offreedom from treatment failure were 93.0% (95% confidence interval [CI], 90.5 to94.8) with the four-cycle ABVD regimen and 91.1% (95% CI, 88.3 to 93.2) with thetwo-cycle regimen. When the effects of 20-Gy and 30-Gy doses of radiation therapy
were compared, there were also no significant differences in freedom from treat-ment failure (P = 1.00) or overall survival (P = 0.61). Adverse events and acute toxiceffects of treatment were most common in the patients who received four cycles ofABVD and 30 Gy of radiation therapy (group 1).
CONCLUSIONS
In patients with early-stage Hodgkins lymphoma and a favorable prognosis, treat-ment with two cycles of ABVD followed by 20 Gy of involved-field radiation therapyis as effective as, and less toxic than, four cycles of ABVD followed by 30 Gy of in-volved-field radiation therapy. Long-term effects of these treatments have not yetbeen fully assessed. (Funded by the Deutsche Krebshilfe and the Swiss FederalGovernment; ClinicalTrials.gov number, NCT00265018.)
The New England Journal of Medicine
Downloaded from www.nejm.org on October 27, 2010. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
-
7/30/2019 Nej Mo a 1000067
2/13
Treatment intensity in ear ly-stage Hodgkins Lymphoma
n engl j med 363;7 nejm.org august 12, 2010 641
Radiation therapy was the originalmainstay of treatment for patients who hadearly-stage Hodgkins lymphoma with a
favorable prognosis. With the use of such tech-niques as extended-field radiation therapy and to-tal lymphoid irradiation, more than 80% of pa-tients with localized disease became long-term
survivors. However, the relapse rate with radia-tion therapy alone ranged from 20 to 40%,1-3 andextended-field radiation therapy and total lymphoidirradiation were associated with the occurrenceof secondary solid tumors.4-8 The integration of achemotherapy regimen consisting of doxorubicin,bleomycin, vinblastine, and dacarbazine (ABVD)9with radiation therapy resulted in greater eff icacyand allowed the radiation field and dose to bereduced, leading to widespread use of the com-bined approaches in patients with early-stageHodgkins lymphoma and a favorable progno-
sis.10,11 Four cycles of ABVD followed by 30 Gy ofinvolved-field radiation therapy is now regardedas the standard of care by many groups.11-14 Theuse of chemotherapy alone has been consideredas a potential alternative approach but remainscontroversial.15-19
Whether the number of treatment cycles andthe radiation dose can be reduced in patients withearly-stage Hodgkins lymphoma and a favorableprognosis remains unclear. In an attempt to re-duce the toxic effects of treatment while retain-ing full control of the cancer, the German Hodg-kin Study Group (GHSG) in 1998 initiated aprospective, randomized, multicenter study (HD10)in which four cycles of ABVD chemotherapy werecompared with two cycles of ABVD, and 30 Gyof involved-field radiation therapy was comparedwith 20 Gy of involved-field radiation therapy inpatients receiving either of the two chemotherapyregimens.
Methods
Study Patients
We enrolled patients who had newly diagnosedHodgkins lymphoma in clinical stage I or II, asconfirmed on histologic examination, with noclinical risk factors. Patients were eligible if theywere between 16 and 75 years of age, had notbeen treated previously for Hodgkins lymphoma,and were free of concurrent disease. (Details re-garding the definitions of clinical risk factors andfull descriptions of the inclusion and exclusioncriteria are provided in the Supplementary Ap-
pendix, available with the full text of this articleat NEJM.org.)
Study Design
HD10 was a multicenter, randomized study offour different treatment regimens in patients withearly-stage Hodgkins lymphoma and a favorable
diagnosis. Patients were recruited and treated at329 hospitals and outpatient practices in Germa-ny, Switzerland, the Netherlands, the Czech Re-public, and Austria. After clinical staging had beencompleted and written informed consent ob-tained, patients were registered at the GHSG cen-tral trial off ice by telephone and then randomlyassigned in a 1:1:1:1 ratio to one of four treat-ment groups: group 1 received four cycles of ABVDfollowed by 30 Gy of involved-field radiation ther-apy; group 2 received four cycles of ABVD fol-lowed by 20 Gy of involved-field radiation thera-
py; group 3 received two cycles of ABVD followedby 30 Gy of involved-field radiation therapy; andgroup 4 received two cycles of ABVD followed by20 Gy of involved-field radiation therapy.
Stratification factors included trial center andprognostic factors that might influence the pri-mary end point, such as age (
-
7/30/2019 Nej Mo a 1000067
3/13
Th e n e w e n g l a n d j o u r n a l o f medicine
n engl j med 363;7 nejm.org august 12, 2010642
uted to the interpretation of the results and vouchfor the accuracy and completeness of the data. TheGHSG chair wrote the first draft of the manuscriptand was supported by the lead statistician. Therewas no commercial involvement in this study and
no financial support from industry.
Chemotherapy
ABVD was administered on days 1 and 15 inmonthly cycles, at the following standard doses:doxorubicin, 25 mg per square meter of body-surface area; bleomycin, 10 mg per square meter;vinblastine, 6 mg per square meter; and dacarba-zine, 375 mg per square meter. If the white-cellcount was less than 2500 per cubic millimeter or
the platelet count was less than 80,000 per cubicmillimeter on a day when chemotherapy was sched-uled to be administered, treatment was postponeduntil normal levels were achieved. Granulocytecolony-stimulating factor was given if clinically
indicated.
Radiation therapy
Before treatment, all sites of disease were definedand documented by the treating medical oncolo-gist and radiation oncologist. A central panel ofexperts in radiation oncology then planned in-volved-field radiation therapy as defined in thestudy protocol according to treatment group and,if necessary, revised the initial staging. The rec-
1370 Patients underwent randomization
346 Were assignedto group 1
(4ABVD + 30 Gy IFRT)
180 Were excluded30 Did not have histologic
features of Hodgkinslymphoma
133 Had wrong initial staging16 Did not meet other inclusion
criteria1 Withdrew before start of
therapy
298 in group 1 wereincluded in analysis
340 Were assignedto group 2
(4ABVD + 20 Gy IFRT)
298 in group 2 wereincluded in analysis
299 in group 4 wereincluded in analysis
295 in group 3 wereincluded in analysis
596 in groups 1 and 2 were included inanalysis for CT comparison (4ABVD)
575 in groups 1 and 3 were included inanalysis for RT comparison (30 Gy)
594 in groups 3 and 4 were included inanalysis for CT comparison (2ABVD)
588 in groups 2 and 4 were included inanalysis for RT comparison (20 Gy)
341 Were assignedto group 3
(2ABVD + 30 Gy IFRT)
343 Were assigned togroup 4
(2ABVD + 20 Gy IFRT)
1190 Were included in analysis for 4-group comparison
1190 Were included in analysis for ABVD comparison
27 Withdrew before RT12 Were in group 16 Were in group 26 Were in group 33 Were in group 4
1163 Were included in analysis for IFRT comparison
Figure 1. Numbers of Patients Randomly Assigned to Treatment Groups and Included in Analyses.
ABVD denotes doxorubicin, bleomycin, vinblastine, and dacarbazine, and IFRT involved-field radiation therapy.
The New England Journal of Medicine
Downloaded from www.nejm.org on October 27, 2010. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
-
7/30/2019 Nej Mo a 1000067
4/13
Treatment intensity in ear ly-stage Hodgkins Lymphoma
n engl j med 363;7 nejm.org august 12, 2010 643
Table1.
BaselineCharacteristicsofthePatientsAccordingtoTreatmentGr
oup.*
Characteristic
T
reatmentGroup
ChemotherapyComparison
Radiatio
nTherapyComparison
AllPatients
(N=1190)
Group1:
4ABVD
+30GyIFRT
(N=298)
Group2:
4AB
VD
+20Gy
IFRT
(N=2
98)
Group3:
2ABVD
+30GyIFRT
(N=295)
Group4:
2ABVD
+20GyIFRT
(N=299)
Groups
1and2
(N=596)
Groups
3and4
(N=594)
Gro
ups
1and3
(N=
575)
Groups
2and4
(N=588)
Agey
r
38.814.3
39.514.2
38.6
14.5
38.314.3
38.714.4
39.014.3
38.514.3
38.613.9
38.614.4
Femalesexn
o.(
%)
465(39.1)
116(38.9)
121(4
0.6)
106(35.9)
122(40.8)
236(39.6
)
228(38.4
)
215(37.4
)
239(40.6
)
AnnArborstagen
o.(
%)
IA
361(30.3)
94(31.5)
92(3
0.9)
99(33.6)
76(25.4)
186(31.2
)
175(29.5
)
188(32.7)
166(28.2
)
IB
24(2.0
)
5(1.7
)
6(2
.0)
7(2.4
)
6(2.0
)
11(1.8
)
13(2.2
)
12(2.1
)
12(2.0
)
IIA
738(62.0)
182(61.1)
184(6
1.7)
175(59.3)
197(65.9)
366(61.4
)
372(62.6
)
346(60.2
)
375(63.8
)
IIB
66(5.5
)
17(5.7
)
15(5
.0)
14(4.7
)
20(6.7
)
32(5.4
)
34(5.7
)
29(5.0
)
35(6.0
)
Infradiaphragmaticdisease
n
o.(
%)
97(8.2
)
25(8.4
)
26(8
.7)
21(7.1
)
25(8.4
)
50(8.4
)
46(7.7
)
44(7.7
)
49(8.3
)
HistologictypeofHodgkins
lymphomano.o
f
patients/totalno.(
%)
Nodularlymphocyte-
predominant
81/1080(7.5
)
20/273(7.3
)
17/263
(6.5
)
24/272(8.8
)
20/272(7.4)
37/535(6.9
)
44/544(8.1
)
43/528
(8.1
)
35/526(6.7
)
Lymphocyte-richclassic
98/1080(9.1
)
23/273(8.4
)
19/263
(7.2
)
35/272(12.9)
21/272(7.7)
42/535(7.9
)
56/544(10.3
)
58/528
(11.0)
40/526(7.6
)
Nodularsclerosing
420/1080(38.9)
101/273(37.0)
112/263
(42.6)
98/272(36.0)
109/272(40.1)213/535(39.8)207/544(38.1
)197/528
(37.3
)
219/526(41.6)
Mixedcellularity
435/1080(40.3)
118/273(43.2)
103/263
(39.2)
99/272(36.4)
115/272(42.3)220/535(41.1)214/544(39.3
)206/528
(39.0)
213/526(40.5)
Lymphocyte-depleted
1/1080(0.1
)
0
0
1/272(0.4
)
0
0
1/544(0.2
)
1/528
(0.2
)
0
Notclassified
45/1080(4.2
)
11/273(4.0
)
12/263
(4.6
)
15/272(5.5
)
7/272(2.6)
23/535(4.3
)
22/544(4.0
)
23/528
(4.4
)
19/526(3.6
)
*PlusminussignsaremeansS
D.
ABVDdenotesdoxorubicin,
bleomycin,vinblastine,anddacarbazine,andIFRTinvolved-fieldradiationtherapy.
Thesegroupsincludealleligiblepatients.
Informationonstageofdisease
wasmissingfor1patientingroup2.
Informationonhistologicsubtypewasavailablefor1080patients(90.7%).
The New England Journal of Medicine
Downloaded from www.nejm.org on October 27, 2010. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
-
7/30/2019 Nej Mo a 1000067
5/13
Th e n e w e n g l a n d j o u r n a l o f medicine
n engl j med 363;7 nejm.org august 12, 2010644
Table2.
AdverseEventsAccordingtoTreatmentGroup.
Event
TreatmentGrou
p*
Chem
otherapyComparison
RadiationTherapyComparison
Group1:
4ABVD
+30GyIFRT
(N=298)
Group2:
4ABVD
+20GyIFRT
(N=298)
G
roup3:
2ABVD
+30GyIFRT
(N
=295)
Group4:
2ABVD
+20GyIFRT
(N=299)
Groups1and2
(N=5
96)
Groups3and4
(N=594)
Groups1and3
(N=575)
Groups2and4
(N=588)
numberofpatients(p
ercent)
no.ofpatients/totalno.(%)
Acutetoxicity(gradeIIIorIV)
Atleastoneevent
304/588
(51.7)
194/585(33.2
)
46/528
(8.7
)
16/553(2.9
)
Anemia
7/588
(1.2
)
1/585(0.2
)
0
0
Thrombopenia
3/588
(0.5
)
0
0
0
Leukopenia
138/588
(23.5)
87/585(14.9
)
0
0
Nauseaorvomiting
79/588
(13.4)
51/585(8.7
)
3/528
(0.6
)
6/553(1.1
)
Mucositis
7/588
(1.2
)
2/585(0.3
)
18/528
(3.4
)
4/553(0.7
)
Gastrointestinaltractdis-
orderordysphagia
11/588
(1.9
)
6/585(1.0
)
30/528
(5.7
)
16/553(2.9
)
Respiratorytractdisorder
12/588
(2.0
)
2/585(0.3
)
2/528
(0.4
)
0
Hairloss
165/588
(28.1)
89/585(15.2
)
6/528
(1.1
)
3/553(0.5
)
Infection
30/588
(5.1
)
10/585(1.7
)
1/528
(0.2
)
0
Pain
9/588
(1.5
)
14/585(2.4
)
3/528
(0.6
)
1/553(0.2
)
Nervoussystemdisorder
12/588
(2.0
)
7/585(1.2
)
2/528
(0.4
)
0
Secondaryneoplasia
Total
17(5.7
)
10(3.4
)
1
4(4.7
)
14(4.7
)
27
(4.5
)
28(4.7
)
31
(5.4
)
24(4.1
)
Acutemyelocyticleuke-
miaormyelodysplastic
syndrome
2(0.7
)
0
0
0
2
(0.3
)
0
2
(0.3
)
0
NonHodgkinslym-
phoma
4(1.3
)
3(1.0
)
3(1.0
)
5(1.7
)
7
(1.2
)
8(1.3
)
7
(1.2
)
8(1.4
)
Solidtumor
11(3.7
)
7(2.3
)
1
1(3.7
)
9(3.0
)
18
(3.0
)
20(3.4
)
22
(3.8
)
16(2.7
)
The New England Journal of Medicine
Downloaded from www.nejm.org on October 27, 2010. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
-
7/30/2019 Nej Mo a 1000067
6/13
Treatment intensity in ear ly-stage Hodgkins Lymphoma
n engl j med 363;7 nejm.org august 12, 2010 645
ommended interval between completion of the ABVDregimen and the start of radiation therapy was 4 to6 weeks. Patients received either 30 Gy or 20 Gy ofinvolved-field radiation therapy in single fractionsof 1.8 to 2.0 Gy administered five times weekly.
Study End Points
The primary eff icacy end point was freedom fromtreatment failure. Overall survival, progression-freesurvival, complete response, and treatment toxicitywere secondary end points. Definitions of the studyend points are provided in the Supplementary Ap-pendix.
Statistical Analysis
Proof of the noninferiority of the less intensive treat-ment, as compared with the standard treatment offour cycles of ABVD plus 30 Gy of involved-fieldradiation therapy, with respect to freedom from
treatment failure at 5 years was the goal for bothchemotherapy and radiation therapy. The noninfe-riority margin was defined as 7% in the study pro-tocol. This led to the following two hypotheses: forchemotherapy, the 5-year rate of freedom from treat-ment failure in the two pooled groups assigned totwo cycles of ABVD would be less than 7% belowthe rate in the two pooled groups assigned to fourcycles, and for radiation therapy, the 5-year rate offreedom from treatment failure in the two pooledgroups assigned to 20 Gy of involved-field radiationtherapy would be less than 7% below the rate in thetwo pooled groups assigned to 30 Gy.
Survival rates for the four groups were comparedwith the use of the KaplanMeier method as well asstratif ied Cox regression analyses for hazard ratios(i.e., the chemotherapy comparison was stratifiedaccording to the radiation therapy assignment andvice versa), whereas outcomes and toxicity rates werecompared with the use of Fishers exact test. Testsof the hypotheses were performed according to theintention-to-treat principle and also on the basis ofthe treatment actually received. Subgroup analyses
were not prespecified in the statistical-analysisplan, but we performed post hoc sensitivity analy-ses that excluded patients with nodular lympho-cyte-predominant Hodgkins lymphoma. The resultsof these sensitivity analyses are presented in theSupplementary Appendix.
In addition, to estimate the combined effect ofreduced chemotherapy and reduced radiation ther-apy, we compared group 1, which received the mostintensive therapy, with group 4, which received the
Death
Totalno.o
fdeaths
15(5.0
)
13(4.4
)
1
6(5.4
)
13(4.3
)
28(4
.7)
29(4.9
)
25(4.3)
22(3.7
)
Causeofdeath
Hodgkinslymphoma
3(1.0
)
2(0.7
)
3(1.0
)
2(0.7
)
5(0
.8)
5(0.8
)
5(0.9)
3(0.5
)
Toxicityofprimary
therapy
3(1.0
)
3(1.0
)
1(0.3
)
0
6(1
.0)
1(0.2
)
1(0.2)
0
Toxicityofsalvage
therapy
0
0
4(1.4
)
1(0.3
)
0
5(0.8
)
3(0.5)
0
Secondaryneoplasia
3(1.0
)
0
5(1.7
)
3(1.0
)
3(0
.5)
8(1.3
)
8(1.4)
3(0.5
)
Cardiovasculardis-
order
3(1.0
)
3(1.0
)
0
3(1.0
)
6(1
.0)
3(0.5
)
3(0.5)
6(1.0
)
*Thesegroupsincludealleligiblepatients.
Informationonthetoxicityofprimarychemotherapy(WorldHealthOrgan
izationgradeIIIorIV)wasavailablefor11
73of1190patients(98.6
%).Informationonthetoxicityofprimary
involved-fieldradiationtherapy
(IFRT)(CommonToxicityCriteriagradeIII
orIV)wasavailablefor1081of1163patie
nts(92.9%).
Atotalof57of1190patientsdied(4.8
%),and55of1190patientshadas
econdaryneoplasm(4.6
%).Medianobservationtimeswere91monthsforoverallsu
rvivaland79monthsfor
freedomfromtreatmentfailure
andprogression-freesurvival;884patients
(74.3%)werefollowedforatleast5years
aftertheendoftherapy.
Causesoftreatment-relateddea
thswerepulmonaryfibrosis,probablybleo
mycin-induced(intwopatientsassignedt
ofourcyclesofABVDandoneassignedto
twocycles);sepsis(in
twopatientsassignedtofourcyclesofABVD);pneumonia(inonepatient
assignedtofourcyclesofABVD);andnot
specified(inonepatientassignedtofour
cyclesofABVD).
The New England Journal of Medicine
Downloaded from www.nejm.org on October 27, 2010. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
-
7/30/2019 Nej Mo a 1000067
7/13
Th e n e w e n g l a n d j o u r n a l o f medicine
n engl j med 363;7 nejm.org august 12, 2010646
least intensive therapy. To detect a possible in-fluence of prognostic factors or interactions be-tween the effects of chemotherapy and those ofradiation therapy, multivariate Cox regressionanalyses were specified in the protocol and per-formed as sensitivity analyses on the same datasets for comparing the two chemotherapy regi-
mens and the two radiation therapy regimens.
Results
Patients
From May 1998 through January 2003, a total of1370 patients were recruited and randomly as-signed to treatment centrally. A total of 180 pa-tients were excluded from all analyses: 30 becausethe reference histologic findings did not confirmHodgkins lymphoma, 133 because of incorrectinitial staging, 16 because they did not meet oth-
er inclusion criteria, and 1 who met the inclusioncriteria but could not subsequently be contacted(Fig. 1).
The baseline characteristics of the study pa-tients are shown in Table 1. No significant dif-ferences were noted among the treatment groupsfor any of the characteristics shown. The medianage of patients at randomization was 36 years(range, 16 to 75), and 61.0% were male; 30.3%had stage IA disease, 2.0% had stage IB, 62.0%had stage IIA, and 5.5% had stage IIB. (Informa-tion on stage of disease was missing for one pa-tient in group 2.) The most frequent subtypediagnosed by the pathology reference panel wasmixed cellularity (40.2%), and 8.1% of patientshad infradiaphragmatic disease.
The main (intention-to-treat) analysis set forthe initial chemotherapy comparison (CT1) com-prised 1190 patients: 596 patients were randomlyassigned to four cycles of ABVD and 594 to twocycles. Of these patients, 36 changed chemother-apy group or had major protocol violations; che-motherapy was not documented for 10 patients.
The per-protocol analysis set for the chemothera-py comparison (CT2) therefore comprised 1144patients (571 randomly assigned to four cycles ofABVD and 573 to two cycles). The main (inten-tion-to-treat) analysis set for the radiation ther-apy comparison (RT1) included 1163 patients:575 patients were randomly assigned to 30 Gy ofinvolved-field radiation therapy and 588 to 20 Gyof involved-field radiation therapy. Of these pa-
tients, 33 had a change in the radiation therapydose or had major protocol violations; radiationtherapy documentation was missing for 17 pa-tients. The per-protocol analysis set for the radia-tion therapy comparison included 1113 patients(557 in the 30-Gy groups and 556 in the 20-Gygroups). There were more protocol violations
and group changes in the groups that received20 Gy of involved-field radiation therapy than inthose that received 30 Gy (P = 0.05). However,since the per protocol analysis and the intention-to-treat analysis had similar results, these im-balances did not affect the f inal conclusions.
Adverse Events
Toxicity of Treatment
Acute toxicity during chemotherapy was more fre-quent in patients who received four cycles ofABVD than in those who received two cycles (Ta-
ble 2). Overall, 51.7% of the patients who receivedfour cycles of ABVD had at least one instance ofsevere toxicity (grade III or IV) as compared with33.2% of those who received two cycles (P
-
7/30/2019 Nej Mo a 1000067
8/13
-
7/30/2019 Nej Mo a 1000067
9/13
Th e n e w e n g l a n d j o u r n a l o f medicine
n engl j med 363;7 nejm.org august 12, 2010648
Deaths
A total of 57 patients (4.8%) died during the fol-low-up period. The most frequent causes of deathwere secondary neoplasia (in 11), Hodgkins lym-phoma (in 10), cardiovascular events (in 9), toxic-ity of primary therapy (in 7), and toxicity of sal-vage therapy (in 5, all after having received two
cycles of ABVD). No difference in mortality wasnoted among the four groups or between the com-bined chemotherapy groups and the combinedradiation therapy groups (Table 2).
Disease Control and Survival
Final treatment outcomes were as follows: 1150of 1190 patients (96.6%) had a complete remission,8 (0.7%) had a partial remission, and 8 (0.7%) didnot have a response (2 had no change and 6 hadprogression of disease during treatment). (Re-sponse criteria are described in the Supplemen-
tary Appendix.) For 24 patients (2.0%), the treat-ment outcome was unclear. The relapse rate was6.0% (71 of 1190 patients). No significant differ-ences were seen in rates of remission, progres-sion, or relapse among the four treatment groupsor between the combined chemotherapy groupsand the combined radiation therapy groups.
The rates of freedom from treatment failure inthe whole intention-to-treat analysis set of 1190patients were estimated to be 92.0% (95% confi-dence interval [CI], 90.2 to 93.5) at 5 years and87.1% (95% CI, 84.5 to 89.3) at 8 years. The over-all survival rates for all 1190 patients were esti-mated to be 96.8% (95% CI, 95.7 to 97.7) at 5 yearsand 94.5% (95% CI, 92.8 to 95.8) at 8 years(Table 3). For the same patients, the rate of pro-gression-free survival was estimated to be 92.4%(95% CI, 90.6 to 93.8) at 5 years and 87.6% (95%CI, 85.0 to 89.7) at 8 years.
Chemotherapy Comparison
In the intention-to-treat analysis, the median ob-servation time for the primary end point, freedom
from treatment failure, was identical in the twochemotherapy groups (79 months). The rate of free-dom from treatment failure at 5 years was 93.0%with four cycles of ABVD (95% CI, 90.5 to 94.8)and 91.1% with two cycles (95% CI, 88.3 to 93.2)(Table 3). On the basis of the stratified Cox re-gression analysis, the hazard ratio for treatmentfailure with two cycles of ABVD as comparedwith four cycles was 1.17 (95% CI, 0.82 to 1.67).The 5-year estimated group difference (two cy-cles vs. four cycles) was 1.9 percentage points
(95% CI, 5.2 to 1.4). The sensitivity analysis,based on treatment received per protocol, showeda 5-year estimated group difference of 2.3 per-centage points (95% CI, 5.6 to 2.9). On the basisof these results, the predefined 7% inferiority oftwo cycles of ABVD plus radiation therapy can beexcluded for the primary end point, freedomfrom treatment failure. The intention-to-treatanalysis showed no significant differences be-tween the two chemotherapy groups for the sec-ondary end points of overall survival (P = 0.93;hazard ratio for death, 1.02 [95% CI, 0.61 to
1.72]) and progression-free survival (P = 0.28;hazard ratio for progression, relapse, or deathfrom any cause, 1.22 [95% CI, 0.85 to 1.77]).
Radiation therapy Comparison
In the intention-to-treat analysis of radiation ther-apy, the median observation time for the primaryend point, freedom from treatment failure, wassimilar in the two groups: 77 months with 20 Gyand 80 months with 30 Gy. The rate of freedomfrom treatment failure at 5 years was 93.4% (95%
Figure 2 (facing page). Freedom from Treatment Failureand Overall Survival.
Two pooled treatment groups were compared with re-spect to chemotherapy regimens (groups 1 and 2 vs.
groups 3 and 4) (Panel A) and radiation therapy doses
(groups 1 and 3 vs. groups 2 and 4) (Panel B). Groups1 and 4 were also compared (Panel C). Group differ-
ences with respect to freedom from treatment failureand overall survival at 5 years were estimated on the
basis of KaplanMeier analyses, and hazard ratios werecalculated with the use of Cox regression analysis (i.e.,
the comparison of the chemotherapy groups was strati-fied according to radiation therapy group, and vice ver-
sa). (For definitions of study end points, see the Sup-plementary Appendix, available with the full text of this
article at NEJM.org.) Data for all patients were analyzedon the basis of the randomly assigned treatment
groups (intention-to-treat principle). Data for overall
survival were censored on the date when the informa-tion was last obtained; when the information lag ex-
ceeded 2 years, data on survival were obtained fromregistries, whenever possible. The median observation
period for freedom from treatment failure was 79months and that for overall survival was 91 months.
The main analysis for the chemotherapy comparisonincluded 1190 eligible patients who received at least
1 dose of the assigned study treatment. According tothe protocol, 27 patients whose disease progressed or
whose chemotherapy was discontinued before the start
of radiation therapy were excluded from the main analy-sis for the radiation therapy comparison. (For methods
and results of the sensitivity analyses, see the Supple-mentary Appendix.)
The New England Journal of Medicine
Downloaded from www.nejm.org on October 27, 2010. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
-
7/30/2019 Nej Mo a 1000067
10/13
Treatment intensity in ear ly-stage Hodgkins Lymphoma
n engl j med 363;7 nejm.org august 12, 2010 649
CI, 91.0 to 95.2) in the 30-Gy group and 92.9%(95% CI, 90.4 to 94.8) in the 20-Gy group (Table 3).The hazard ratio for treatment failure with 20 Gy
as compared with 30 Gy was 1.00 (95% CI, 0.68 to1.47). The 5-year estimated group difference (20 Gyvs. 30 Gy) was 0.5 percentage points (95% CI,
FreedomfromT
reatment
Failure(%)
100
80
90
70
60
4030
10
50
20
00 12 24 36 48 120
Months
B Radiation Therapy Comparison
A Chemotherapy Comparison
No. ofPatientsat Risk
4ABVD2ABVD
596594
554555
532530
506498
479473
60
430410
72
330314
84
226225
96
131131
108
5754
69
OverallSurvival(%)
100
80
90
70
60
4030
10
50
20
00 12 24 36 48 120
MonthsNo. of
Patientsat Risk
4ABVD2ABVD
596594
583589
575578
569572
562567
60
541549
72
471482
84
348361
96
227239
108
130126
2436
Difference at 5 yr, 1.9 percentage points (95% CI, 5.2 to 1.4)Hazard ratio, 1.17 (95% CI, 0.82 to 1.67)
Difference at 5 yr, 0.5 percentage points (95% CI, 2.6 to 1.6)Hazard ratio, 1.02 (95% CI, 0.61 to 1.72)
Freedomf
romT
reatment
Failure(%)
100
8090
70
60
40
30
10
50
20
00 12 24 36 48 120
MonthsNo. of
Patientsat Risk
30 Gy IFRT20 Gy IFRT 575588 553550 526531 499502 471478
60
426411
72
328314
84
235215
96
139123
108
6150 87
OverallSurvival(%)
100
8090
70
60
40
30
10
50
20
00 12 24 36 48 120
MonthsNo. of
Patientsat Risk
30 Gy IFRT20 Gy IFRT 575588 570583 561575 556568 552560
60
535539
72
469468
84
352346
96
228232
108
125131 3228
Difference at 5 yr, 0.5 percentage points (95% CI, 3.6 to 2.6)Hazard ratio, 1.00 (95% CI, 0.68 to 1.47)
Difference at 5 yr, 0.2 percentage points (95% CI, 2.0 to 1.7)Hazard ratio, 0.86 (95% CI, 0.49 to 1.53)
C Comparison of Groups 1 and 4
Freedomf
romT
reatment
Failure(%)
100
80
90
70
60
40
30
10
50
20
0 0 12 24 36 48 120
MonthsNo. of
Patientsat Risk
Group 1Group 4
298299
277275
264265
255252
239239
60
217199
72
167151
84
121110
96
7466
108
3528
34
OverallSurvival(%)
100
80
90
70
60
40
30
10
50
20
0 0 12 24 36 48 120
MonthsNo. of
Patientsat Risk
Group 1Group 4
298299
293298
289293
286289
283285
60
271273
72
240241
84
182182
96
116122
108
6364
1216
Difference at 5 yr, 1.6 percentage points (95% CI, 6.3 to 3.1)Hazard ratio, 1.07 (95% CI, 0.65 to 1. 77)
Difference at 5 yr, 0.4 percentage points (95% CI, 3.4 to 2.7)Hazard ratio, 0.85 (95% CI, 0.41 to 1. 79)
4ABVD (groups 1 and 2)
2ABVD (groups 3 and 4)
30 Gy IFRT (groups 1 and 3)
20 Gy IFRT (groups 2 and 4)
4ABVD (groups 1 and 2)
2ABVD (groups 3 and 4)
30 Gy IFRT (groups 1 and 3)
20 Gy IFRT (groups 2 and 4)
Group 1
Group 4
Group 1
Group 4
The New England Journal of Medicine
Downloaded from www.nejm.org on October 27, 2010. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
-
7/30/2019 Nej Mo a 1000067
11/13
Th e n e w e n g l a n d j o u r n a l o f medicine
n engl j med 363;7 nejm.org august 12, 2010650
3.6 to 2.6). The sensitivity analysis based on ther-apy received showed a 5-year estimated group dif-ference of 0.2 percentage points (95% CI, 3.3 to2.8). Thus, the predefined 7% inferiority of che-motherapy plus 20 Gy of radiation therapy can beexcluded for the primary end point (freedom fromtreatment failure). The intention-to-treat analysis
showed no significant differences between theradiation therapy groups for the secondary endpoints of overall survival (P = 0.61; hazard ratiofor death, 0.86 [95% CI, 0.49 to 1.53]) and pro-gression-free survival (P = 0.98; hazard ratio forprogression, relapse, or death from any cause, 1.01[95% CI, 0.68 to 1.48]). Overall, the rates of free-dom from treatment failure might appear to behigher than those in a pure intention-to-treatanalysis, since patients who dropped out beforeradiation therapy were excluded from this analy-sis. However, this was unlikely to affect between-
group comparisons.
Prespecified Regression Analyses
Prespecified factors included in the multivariatemodel were age above 50 years (P
-
7/30/2019 Nej Mo a 1000067
12/13
Treatment intensity in ear ly-stage Hodgkins Lymphoma
n engl j med 363;7 nejm.org august 12, 2010 651
follow-up is needed to identify differences in long-term toxicity, such as secondary neoplasia andsevere organ damage, among different treatmentapproaches. Given that many of the late, fatal com-plications of radiation therapy do not emerge untilthe second decade after treatment, our data can-not speak to the effect of treatment on overall
survival.Since radiation therapy is associated with the
development of secondary solid tumors 5 to 25years after initial treatment,4-8 some groups ad-vocate the use of chemotherapy alone for patientswith early-stage Hodgkins lymphoma. Usually, sixcycles of ABVD are given, and there has been somecontroversy on this issue.15-19 For this group ofpatients, combined-approach treatment programshave provided superior tumor control when com-pared directly with chemotherapy alone in somestudies20-24 but not in others.15,19 Currently, com-
bined-approach treatment programs are widelyused as the treatment of choice in early-stageHodgkins lymphoma, and our study suggests thata shorter chemotherapy regimen with a lowerradiation dose preserves a high level of diseasecontrol. With an overall survival rate of 95.1% at8 years, some patients may still be overtreated.
However, the established clinical risk factors,which are based on measures such as the Inter-national Prognostic Score,25 currently do not allowidentification of patients who can be cured witheven less treatment. The use of positron-emission
tomography (PET) might help to discriminate be-tween patients at low risk and those at high risk,both early in the course of chemotherapy26 andafter its completion.27 The potential effect of PETin patients with Hodgkins lymphoma has alsobeen suggested in a number of retrospective, non-randomized studies.28-30 Several ongoing trials
are evaluating the role of PET in identifying pa-tients with early Hodgkins lymphoma and a fa-vorable prognosis who might not need additionalradiation therapy after two cycles of ABVD (theGerman Hodgkin Study Group Hodgkin Disease16 [the current GHSG HD16] trial [ClinicalTrials.gov number, NCT00736320]) or after three cy-cles of ABVD (the European Organization for Re-search and Treatment of Cancer [EORTC H10F]trial [ClinicalTrials.gov number, NCT00433433]and others).
In summary, the HD10 trial showed that in
patients with early-stage Hodgkins lymphoma anda favorable prognosis, treatment with two cyclesof ABVD followed by 20 Gy of involved-field ra-diation therapy is as effective as, and less toxicthan, four cycles of ABVD followed by 30 Gy ofinvolved-field radiation therapy.
Supported by grants from the Deutsche Krebshilfe and theSwiss Federal Government.
Disclosure forms provided by the authors are available withthe full text of this article at NEJM.org.
We thank Hiltrud Nisters-Backes, Bettina Koch, HannoloreOssadnik, Dagmar Bhmer, Thomas Schober, and Marina Schu-macher for their technical support and Dr. Joachim Yahalom forhis critical review of the draft manuscript.
appendix
The authors affiliations are as follows: the Department of Internal Medicine (A.E., P.B.), the German Hodgkin Study Group, Departmentof Internal Medicine I (A.E., A.P., P.B., M.F., V.D.), and the Department of Radiation Oncology (H.T.E., R.-P.M.), University of Cologne,Cologne; the Department of Hematology and Oncology, Charit-Universittsmedizin Berlin, Berlin (B.D.); the Departments of RadiationOncology (B.B.) and Internal Medicine II (M.S.), University of Tbingen, Tbingen; Pius-Hospital Oldenburg, Klinik fr Strahlenthera-pie und internistische Onkologie, Oldenburg (K.C.W.); Klinikum Nrnberg Nord, Medizinische Klinik 5, Schwerpunkt HmatologieOnkologie, Nrnberg (M.W.); the Departments of Radiation Oncology (J.D.) and Internal Medicine V (A.H.), University of Heidelberg,Heidelberg; the Department of Radiotherapy, Rheinisch-Westflische Technische Hochschule Aachen, Aachen (M.J.E.); Ludwig-Maxi-milians University, University Hospital Grosshadern, Department of Internal Medicine III, Munich (A.R.); the Department of Hematol-ogy, Hemostasis, Oncology, and Stem-Cell Transplantation, Hannover Medical School, Hannover (A.G.); the Department of Hematol-ogy and Oncology, Georg-August University Gttingen, Gttingen (L.T.); the Department of Oncology, Hematology, and Bone MarrowTransplantation, University Medical Center Hamburg, Hamburg (C.B.); the Department of HematologyOncology, Siloah Clinic, Han-nover (H.K.); Evangelische Krankenhaus Hamm, Medizinische Klinik, Abteilung fr Hmato-Onkologie, Hamm (J.S.); and the Institute
of Pathology, University of Wrzburg, Wrzburg (H.-K.M.-H.) all in Germany; Swiss Group for Clinical Cancer Research, Bern,Switzerland (A.L.); Medical Department III, University of Salzburg, Salzburg, Austria (R.G.); and the Department of Internal MedicineHematooncology, University Hospital and Faculty of Medicine of Masaryk University, Brno, Czech Republic (Z.K.).
References
Mauch PM, Armitage JO, Diehl V, et1.al., eds. Hodgkins disease. Philadelphia:Lippincott Williams & Wilkins, 1999.
Carde P, Burgers JM, Henry-Amar M,2.et al. Clinical stages I and II Hodgkinsdisease: a specifically tailored therapy ac-cording to prognostic factors. J Clin On-col 1988;6:239-52.
Specht L, Gray RG, Clarke MJ, Peto R.3.Influence of more extensive radiotherapyand adjuvant chemotherapy on long-termoutcome of early-stage Hodgkins disease:a meta-analysis of 23 randomised trialsinvolving 3,888 patients. J Clin Oncol 1998;16:830-43.
Boivin JF, Hutchinson GB, Zauber AG,4.
et al. Incidence of second cancers in pa-tients treated for Hodgkins disease. J NatlCancer Inst 1995;87:732-41.
van Leeuwen FE, Klokman WJ, vant5.Veer MB, et al. Long-term risk of secondmalignancy in survivors of Hodgkins dis-ease treated during adolescence or youngadulthood. J Clin Oncol 2000;18:487-97.
The New England Journal of Medicine
Downloaded from www.nejm.org on October 27, 2010. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.
-
7/30/2019 Nej Mo a 1000067
13/13
n engl j med 363;7 nejm.org august 12, 2010652
Treatment intensity in ear ly-stage Hodgkins Lymphoma
Ng AK, Bernardo MP, Weller E, et al.6.Long-term survival and competing causesof death in patients with early-stage Hodg-kins disease treated at age 50 or younger.J Clin Oncol 2002;20:2101-8.
Specht L. Very long-term follow-up of7.the Danish National Hodgkin StudyGroups randomized tr ial of radiotherapy(RT) alone vs. combined modality treat-
ment (CMT) for early stage Hodgkin lym-phoma, with special reference to secondtumors and overall survival. Blood 2003;102:Suppl:2351a. abstract.
Franklin J, Pluetschow A, Paus M, et8.al. Second malignancy risk associatedwith treatment of Hodgkins lymphoma:meta-analysis of the randomized trials.Ann Oncol 2006;17:1749-60.
Bonadonna G, Zucali R, Monfardini9.S, De Lena M, Uslenghi C. Combinationchemotherapy of Hodgkins disease withadriamycin, bleomycin, vinblastine, imi-dazole carboxamide versus MOPP. Cancer1975;36:252-9.
Engert A, Franklin J, Eich HT, et al.10.
Two cycles of doxorubicin, bleomycin,vinblastine, and dacarbazine plus extended-field radiotherapy is superior to radio-therapy alone in early favourable Hodgkinslymphoma: final results of the GHSG HD7Trial. J Clin Oncol 2007;25:3495-502.
Ferm C, Eghbali H, Meerwaldt JH, et11.al. Chemotherapy plus involved-field ra-diation in early-stage Hodgkins disease.N Engl J Med 2007;357:1916-27.
Noordijk EM, Carde P, Dupouy N, et12.al. Combined-modality therapy for clini-cal stage I or II Hodgkins lymphoma:long-term results of the European Organ-isation for Research and Treatment ofCancer H7 randomized controlled trials.
J Clin Oncol 2006;24:3128-35.Engert A, Schiller P, Josting A, et al.13.
Involved-field radiotherapy is equally effec-tive and less toxic as compared with ex-tended-field radiotherapy after four cyclesof chemotherapy in patients with early-stage unfavourable Hodgkins lymphoma:results of the HD8 trial of the GermanHodgkins Lymphoma Study Group(GHSG). J Clin Oncol 2003;21:3601-8.
NCCN clinical practice guidelines in14.oncology: Hodgkin disease/lymphoma.Fort Washington, PA: National Compre-hensive Cancer Network, 2008. (Availableat http://www.nccn.org.)
Straus DJ, Portlock CS, Oin J, et al.15.Results of a prospective randomized clin-ical trial of doxorubicin, bleomycin, vin-blastine, and dacarbazine (ABVD) followed
by radiation therapy (RT) versus ABVDalone for stages I, II and IIIA non-bulkyHodgkin disease. Blood 2004;104:3483-9.
Connors JM. State-of-the-art therapeu-16.tics: Hodgkins lymphoma. J Clin Oncol2005;23:6400-8.
Canellos GP. Chemotherapy alone for17.early Hodgkins lymphoma: an emergingoption. J Clin Oncol 2005;23:4574-6.
Yahalom J. Dont throw out the baby18.with the bathwater: on optimizing cureand reducing toxicity in Hodgkins lym-phoma. J Clin Oncol 2006;24:544-8.
Nachman JB, Posto R, Herzog P, et al.19.Randomized comparison of low-dose in-volved-field radiotherapy and no radio-
therapy for children with Hodgkins dis-ease who achieve a complete response tochemotherapy. J Clin Oncol 2002;20:3765-71.
Bloomfield CD, Pajak TF, Glicksman20.AS, et al. Chemotherapy and combinedmodality therapy for Hodgkins disease: aprogress report on Cancer and LeukemiaGroup B studies. Cancer Treat Rep 1982;66:835-46.
Pavlovsky S, Maschio M, Santarelli21.MT, et al. Randomized trial of chemo-therapy versus chemotherapy plus radio-therapy for stage I-II Hodgkins disease.J Natl Cancer Inst 1988;80:1466-73.
Aviles A, Delgado S. A prospective22.
clinical trial comparing chemotherapy,radiotherapy and combined therapy in thetreatment of early stage Hodgkins dis-ease with bulky disease. Clin Lab Haema-tol 1998;20:95-9.
Meyer RM, Gospodarowicz MK, Con-23.nors JM, et al. Randomized comparisonof ABVD chemotherapy with a strategythat includes radiation therapy in patientswith limited-stage Hodgkins lymphoma:
National Cancer Institute of Canada Clin-ical Trials Group and the Eastern Coop-erative Oncology Group. J Clin Oncol2005;23:4634-42.
Eghbali H, Brice P, Creemers GY, et al.24.Comparison of three radiation dose levelsafter EBVP regimen in favourable supra-diaphragmatic clinical stages (CS) IIIHodgkins lymphoma (HL): preliminary
results of the EORTC-GELA H9-F Trial.Blood 2005;106:Suppl:814a. abstract.Hasenclever D, Diehl V. A prognostic25.
score for advanced Hodgkins disease.N Engl J Med 1998;339:1506-14.
Hutchings M, Loft A, Hansen M, et al.26.FDG-PET after two cycles of chemothera-py predicts treatment failure and progres-sion-free survival in Hodgkin lymphoma.Blood 2006;107:52-9.
Kobe C, Dietlein M, Franklin J, et al.27.Positron emission tomography has a highnegative predictive value for progressionor early relapse for patients with residualdisease after first-line chemotherapy inadvanced-stage Hodgkin lymphoma. Blood
2008;112:3989-94.Spaepen K, Stroobants S, Dupont P, et28.
al. Can positron emission tomographywith [18F]-f luorodeoxyglucose after first-line treatment distinguish Hodgkins dis-ease patients who need additional therapyfrom others in whom additional therapywould mean avoidable toxicit y? Br J Hae-matol 2001;115:272-8.
Zinzani PL, Tani M, Fanti S, et al.29.Early positron emission tomography (PET)restaging: a predictive final response inHodgkins disease patients. Ann Oncol2006;17:1296-300.
Gallamini A, Hutchings M, Rigacci L,30.et al. Early interim 2-[18F]-fluoro-2-
deoxy-D-glucose positron emission to-mography is prognostically superior tointernational prognostic score in advanced-stage Hodgkins lymphoma: a report froma joint Italian-Danish study. J Clin Oncol2007;25:3746-52.Copyright 2010 Massachusetts Medical Society.
earlyjobalertserviceavailableatthenejmcareercenter
Register to receive weekly e-mail messages with the latest job openingsthat match your specialty, as well as preferred geographic region,
practice setting, call schedule, and more. Visit the NEJM CareerCenteratNEJMjobs.orgfor more information.
The New England Journal of Medicine