orbital tumors in usa: difference in survival patterns
TRANSCRIPT
Cancer Epidemiology 38 (2014) 515–522
Orbital tumors in USA: Difference in survival patterns
Waleed M. Hassan a, Ahmad S. Alfaar a,b,c,1,*, Mohamed S. Bakry a, Sameera Ezzat a,d
a Department of Research, Children’s Cancer Hospital Egypt 57357, Cairo, Egyptb Cairo University School of Medicine, Cairo, Egyptc Ulm University, 89069 Ulm, Baden-Wurttemberg, Germanyd National Liver Institute, Menoufiya University, Shebin Elkom, Egypt
A R T I C L E I N F O
Article history:
Received 11 November 2013
Received in revised form 28 June 2014
Accepted 1 July 2014
Available online 20 July 2014
Keywords:
SEER
Orbit
Lymphoma
Carcinoma
Rhabdomyosarcoma
Melanoma
Survival
A B S T R A C T
Introduction: There is a wide range of tumors affecting the orbital adnexa. Key such tumors include
lymphomas, carcinomas, melanomas and rhabdomyosarcomas. Several studies have proposed that
these histological subtypes differ in their survival outcomes. In this study we aim to describe the
difference in survival outcomes between such subtypes. Methods: The SEER database was used to gather
patient information. All 18 SEER registries were used. Patients diagnosed from 1996 to 2005 were
included in the analysis. Observed five-year survival rate was calculated using the SEER*Stat software
version 8.1.2. Data were extracted into IBM SPSS version 20 to generate Kaplan Meier curve for each
group. Results: There were 2180 patients in the SEER databases who met the selection criteria.
Lymphomas were the most common histology in adults. The overall five-year observed survival for all
lymphoma patients was 75.9% (95% CI: 73.7–78.1). There was statistically significant difference between
observed survival rates of lymphoma subtypes. Carcinomas were the second most common tumors.
Their five-year observed survival rate in our study was 60.4%. There was no statistically significant
difference between carcinoma subtypes’ observed survival rates in the 20–49 age group, while, in the
older age group, the difference was found to be statistically significant. Rhabdomyosarcomas were the
most common tumors in children. The overall five-year observed survival rate for rhabdomyosarcomas
patients was 89.8%. There was no statistically significant difference between observed survival rates of
rhabdomyosarcomas subtypes. There was no statistically significant difference between relative survival
rates according to gender and treatment received except within melanomas. Conclusion: In adults,
lymphomas have better survival rates than carcinomas. Whereas the lymphoma subtype can be used as a
determinant prognostic factor in any age, the carcinoma subtype can be used as such a determinant in
older age groups only. In children, rhabdomyosarcomas are the predominant tumors affecting the orbital
adnexa. Further studies are needed to determine if the difference between embryonal rhabdomyosar-
coma and alveolar rhabdomyosarcoma observed survival rates are statistically significant.
� 2014 Elsevier Ltd. All rights reserved.
Contents lists available at ScienceDirect
Cancer EpidemiologyThe International Journal of Cancer Epidemiology, Detection, and Prevention
jou r nal h o mep age: w ww.c an cer ep idem io log y.n et
1. Introduction
Tumors that occur in the orbital region are either involvingintra-ocular layers or extra-ocular orbital soft-tissue structures.Whereas all affect the orbital region, only adnexal tumors are
* Corresponding author at: Research Department, Children’s Cancer Hospital
Egypt 57357, 1 Seket Al-Emam Street – El-Madbah El-Kadeem Yard – El-Saida
Zeinab, Cairo, Egypt. Tel.: +20 1001557426; fax: +20 226178046.
E-mail addresses: [email protected] (W.M. Hassan),
[email protected], [email protected] (A.S. Alfaar),
[email protected] (M.S. Bakry), [email protected] (S. Ezzat).1 Current address: Ulm University, 89069 Ulm, Baden-Wurttemberg, Germany.
http://dx.doi.org/10.1016/j.canep.2014.07.001
1877-7821/� 2014 Elsevier Ltd. All rights reserved.
named after the orbit. The most common primary intra-oculartumors include retinoblastomas in childhood and choroidalmelanomas in adults, in addition to other rarer entities. Orbitalmasses comprise a wide variety of lesions depending on its site oforigin.
Based on their behavior, these lesions differ in their manage-ment. Most are benign, as hemangiomas, adenoid cysts, lacrimalgland adenomas and benign skin lesions [1,2]. Others, however, aremalignant and may require special treatment, as lymphomas,carcinomas and rhabdomyosarcomas. In most of these cases it isthe ophthalmologist that is first to examine the lesions. Survivalrates of these malignant tumors have not been studied before in asingle study. In this study we review the prognosis of orbital
W.M. Hassan et al. / Cancer Epidemiology 38 (2014) 515–522516
tumors (focusing on lymphomas, carcinomas, melanomas andrhabdomyosarcomas, since they are the most common orbitalaffecting tumors). The study focuses on survival rates and responseto treatment in population-based settings. We aimed to comparesurvival of different histology subtypes, and were able to identifywhich histology subtypes could be used as a determinantprognostic factor. Moreover, response to treatment was calculatedusing survival rates after receiving different treatment modalitieseither surgery or radiological. It should be noted that properevaluation of different treatment modalities is better evaluatedusing randomized controlled clinical trials. As such, the popula-tion-based results regarding treatment presented here should beinterpreted with caution. Unfortunately, data on chemotherapytreatment was not available, so treatment was only classified intoradiation alone, surgery alone, or both.
The study was built based on the SEER 18 population basedcancer registries. SEER has been collecting patients’ informationsince 1973. Survival rates have improved since 1973 due toadvances in treatment modalities [3]. Therefore, we includedpatients from 1996 to 2005 to be able to describe survival in thelight of these advances. We describe observed survival rates, andwe use them to compare differences in survival between differenthistological subtypes. Observed survival rates are chosen to bepresented in this paper to provide clinically relevant reference forclinicians. Relative survival rates are also presented in Table 2, andthey were used to assess differences in survival regarding sex, raceand treatment [4].
2. Methods
SEER databases were used to gather patients’ data. All 18 SEERregistries were used. The SEER 18 registries consist of Atlanta,Connecticut, Detroit, Hawaii, Iowa, New Mexico, San Francisco-Oakland, Seattle-Puget Sound, Utah, Los Angeles, San Jose-Monterey, Rural Georgia, the Alaska Native Tumor Registry,Greater California, Kentucky, Louisiana, New Jersey, and GreaterGeorgia. SEER*Stat software version 8.1.2 was used to extractpatients’ data [5].
Patients were identified as having orbital tumors using the ICD-O 3 topography codes C69.0-Conjunctiva, C69.5-Lacrimal glandand C69.6-Orbit, not otherwise specified (NOS) [6]. Only patientsdiagnosed from 1996 till 2005 with known age and malignantbehavior were included in the analysis cohort. Patients’ data wasgathered up to December 2010. Patients diagnosed between 2005and 2010 were available in the SEER database but were notincluded in our analysis to ensure the availability of five-yearsurvival follow-up of the cohort.
Patients were then divided into 3 age groups (0–19, 20–49,�50). Each age group was further stratified by histology subtypeusing ICD-O 3 morphology codes. Due to the diversity in histologysubtypes, patients with rare histology subtypes in each age groupwere excluded from the cohort analysis. A cutoff for rare histologysubtypes in the 20–49 age group was set at 10 patients, while in�50 age group the cutoff was determined as 20. SupplementaryTable 1 shows the count of excluded histology subtypes.
See Supplementary Table 1 as supplementary file. Supplemen-tary material related to this article can be found, in the onlineversion, at http://dx.doi.org/10.1016/j.canep.2014.07.001.
Data were then entered in IBM SPSS version 20 to generateKaplan Meier curves for each subgroup. All comparisons betweenobserved survival curves were done using the log rank test. Logrank chi-square statistics were generated and p-values werecalculated. SEER*Stat 8.1.2 was used to check for statisticalsignificance using the Z-test in five-year relative survival ratesbetween different treatment modalities received (radiation,surgery or both), race and gender for each subgroup. The Z-test
used by SEER*Stat compares the survival curves of two groups ofcases up to a selected survival duration point, Five-year relativesurvival was the survival duration point chosen for comparison,and Z values more than 1.96 were considered statisticallysignificant.
3. Results
The SEER database included 2180 patients who met theselection criteria. A summary of patients’ demographic data(gender, age and treatment received) is displayed in Table 1.The overall five-year observed survival was 72.2% (95% CI: 70.2–74.1). In the 0–19 age group there were 59 patients with an overallfive-year observed survival rate of 89.8% (95% CI: 82.0–97.6). The20–49 age group (380 patients) had a comparable five-yearobserved survival rate of 92.0% (95% CI: 89.3–94.7). The �50 agegroup had the highest number of patients (1741 patient) with thelowest five-year observed survival rate of 67.0% (95% CI: 64.8–69.1) (data not shown).
The 20–49 age group was subdivided into 3 subgroups:lymphomas, melanomas and carcinomas. The �50 age groupwas subdivided into: lymphomas – with each histology subtypehaving more than 72 patients; lymphomas – with each histologysubtype having less than 72 patients; carcinomas and melanomas.Carcinomas had lower observed survival rates than lymphomas.Lymphomas were separated into two groups based on the mediannumber of patients within different histology subtypes for the sakeof clear visual representation. Rhabdomyosarcomas had thehighest observed survival rates. Kaplan Meier for each subgroupobserved survival is shown in Figs. 1–4. Results for the log ranktest, chi square, p value are shown below each curve.
3.1. Lymphomas
Lymphomas comprised 66.1% of the total number of cases. Inthe 20–49 age group there were five common subtypes whileolder age group tumors were classified into ten histologicalsubtypes. Statistical significance in observed survival rates wasnoted for lymphomas in the 20–49 age group (p value < 0.001),and for lymphomas �50 age group with N � 50. Conversely,lymphomas in the �50 age group with N < 72 showed nostatistically significant difference in observed survival rates. Themost common subtypes were extra-nodal marginal zonelymphomas of mucosal-associated lymphoid tissue (MALT) anddiffuse large B-cell lymphomas (DLBCL). MALT had bettersurvival rates than DLBCL. The overall five-year observed survivalfor all lymphoma patients was 75.9% (95% CI: 73.7–78.1), whilethe relative survival was 90.7% (95% CI: 87.7–93.0). Fig. 1 showsKaplan Meier curves for observed survival of lymphoma patients.No statistically significant difference was detected betweenrelative survival of patients with ocular lymphoma with regardsto sex, race or modality of treatment.
3.2. Carcinomas
Carcinomas were 22.0% of the total number of cases. In the 20–49 age group there were only two common subtypes. While in the�50 age group there were five common subtypes. There was nostatistical significance in observed survival rates between carci-nomas subtypes models in the 20–49 age group, while in the olderage group there was a statistically significant difference betweenobserved survival rates of carcinoma subtypes (p value < 0.001).The five-year observed survival rate for all carcinoma patients inour study was 60.4% (95% CI: 55.9–64.9), while the relative survivalwas 74.8% (95% CI: 68.6–80.0). Fig. 2 shows Kaplan Meier curves forobserved survival of carcinoma patients.
Table 1Demographics of patients included in the study in the period 1996–2005 from the SEER 18 registries which consist of Atlanta, Connecticut, Detroit, Hawaii, Iowa, New Mexico,
San Francisco-Oakland, Seattle-Puget Sound, Utah, Los Angeles, San Jose-Monterey, Rural Georgia, the Alaska Native Tumor Registry, Greater California, Kentucky, Louisiana,
New Jersey, and Greater Georgia.
Histology subtype Age group Gender Treatment
0–19 years 20–49 years �50 years Male Female Radiation Surgery Unknown
Carcinomas
8010/3: Carcinoma, NOS 28 17 11 7 9 12
8070/3: Squamous cell carcinoma, NOS 64 305 283 86 38 253 78
8071/3: Squamous cell carcinoma, keratinizing, NOS 31 22 9 6 16 9
8140/3: Adenocarcinoma, NOS 12 9 3 7 3 2
8200/3: Adenoid cystic carcinoma 17 22 20 19 24 12 3
Melanoma
8720/3: Malignant melanoma, NOS 46 154 89 111 21 135 44
Rhabdomyosarcomas
8900/3: Rhabdomyosarcoma, NOS 7 4 3 4 2 1
8910/3: Embryonal rhabdomyosarcoma 44 29 15 35 3 6
8920/3: Alveolar rhabdomyosarcoma 8 3 5 8
Lymphomas
9590/3: Malignant lymphoma, NOS 24 93 46 71 68 7 42
9591/3: Malignant lymphoma, non-Hodgkin, NOS 28 101 57 72 77 21 31
9670/3: Malignant lymphoma, small B lymphocytes,
NOS [OBS 2012+] see 9823/3
17 101 48 70 76 22 20
9671/3: Lymphoplasmacytic lymphoma (NHL) 27 14 13 15 5 7
9673/3: Mantle cell lymphoma 57 39 18 30 8 19
9680/3: Diffuse large B-cell (NHL) lymphoma (DLBCL) 14 131 63 82 79 24 42
9690/3: Follicular lymphoma, NOS 37 15 22 19 7 11
9691/3: Follicular lymphoma, grade 2 36 16 20 15 8 13
9695/3: Follicular lymphoma, grade 1 51 16 35 31 10 10
9699/3: Extranodal marginal zone lymphoma of
mucosal-assoc. lymphoid tissue-MALT
170 555 294 431 462 101 162
W.M. Hassan et al. / Cancer Epidemiology 38 (2014) 515–522 517
No statistically significant difference was detected betweenrelative survival of ocular carcinoma patients with regards to sex orrace. Surgery has shown significantly higher relative survival rates(79.7%, 95% CI: 72.5–85.2) when compared to radiation (56.4%, 95%CI: 42.2–68.5, Z = 3.552) and radiation with radiotherapy (65.3%,95% CI: 47.6–78.2, Z = 2.261), as shown in Table 2.
3.3. Melanomas
Melanomas were 9.2% of the total cases. In the 20–49 age groupand the �50 age group there was only one major subtype 8720/3:malignant melanoma, NOS. In the 20–49 age group, the five-yearobserved survival rate was 81.8% (95% CI: 70.4–93.2), while in the�50 age group it was 60% (95% CI: 52.1–67.8). The five-yearobserved survival rate for all melanoma patients in this study was64.9% (95% CI: 58.2–71.6), while the relative survival rate was75.6% (95% CI: 66.9–82.4). Fig. 3 shows Kaplan Meier curves forobserved survival of melanoma patients.
Fig. 1. Observed survival rates for lymphoma patients. (A) Observed survival for differen
lymphoma subtypes occurring in more than 72 patients for age group � 50 years. (C) Obse
age group � 50 years.
No statistically significant difference was detected betweenrelative survival of ocular melanoma patients with regards to sexor race. Surgery has shown significantly higher relative survivalrates (80%, 95% CI: 69.3–87.3) compared to radiation (52.7%, 95%CI: 27.5–72.8, Z = 2.664) and radiation with radiotherapy (31.5%,95% CI: 3.8–66.5, Z = 3.387) as shown in Table 2.
3.4. Rhabdomyosarcomas
Rhabdomyosarcomas affected the 0–19 age group only. Theyrepresented 2.7% of the total number of cases. Rhabdomyosarcomaswere divided in this study into three subtypes. Embryonalrhabdomyosarcoma was the most common subtype. They had highsurvival rates with no statistically significant difference between thesubtypes. The overall five-year observed survival rate for rhabdo-myosarcomas patients was 89.8% (95% CI: 82–97.6), while therelative survival was 93.1% (95% CI: 79.7–97.7). Fig. 4 shows KaplanMeier curve for observed survival of rhabdomyosarcoma patients.
t lymphoma subtypes for the age group 20–49. (B) Observed survival for different
rved survival for different lymphoma subtypes occurring in less than 72 patients for
Fig. 2. Observed survival rates for carcinoma patients. (A) Observed survival rates for different carcinoma subtypes for the age group 20–49. (B) Observed survival rates for
different carcinoma subtypes for the age group � 50 years.
W.M. Hassan et al. / Cancer Epidemiology 38 (2014) 515–522518
No statistically significant difference was detected betweenrelative survival of ocular rhabdomyosarcoma patients withregards to sex, race or treatment modality.
4. Discussion
Orbital tumors include a wide scope of hamartomas – especiallyvascular ones – which are the most common orbital tumors.However, due to its benign nature, simple excision is the ultimatesolution in most of these lesions with excellent outcomes. When itcomes to malignant tumors, physicians eye the probability ofmisdiagnosing a malignant lesion as benign, and patients ask abouttheir survival and long term sequelae. Crucially, survival rates varybetween different types of malignant tumors of the orbit.
Lymphomas are the predominant cancer affecting orbital andlacrimal gland sites in adults [7,8]. The five-year observed survivalin this study (75.9%) was similar to that stated in similar studies[9]. Lymphomas subtypes are more common in females except formantle cell lymphomas [10,11]. A statistically significant differ-ence was found in survival between different lymphoma subtypesin the 20–49 age group and the �50 age group where each subtype
Fig. 3. Observed survival rates for melanoma patients. (A) Observed survival rates for
group � 50 years.
affected more than 72 patients. There was no statistical signifi-cance in survival difference between different subtypes in the �50age group where each subtype affected < 72 patients (Fig. 1). Inthese tumors (MALT) was the predominant subtype [7,8,12–14],while (DLBCL) is the second most common tumor [15]. In our studypopulation MALT patients had higher significant survival ratesthan DLBCL. These results are consistent with similar studies madeon smaller samples of patients [13,14,16–20]. The high survivalrate of MALT patients has reached an extent that a study on 36MALT patients proposed that they may not require initialtreatment and still have similar outcomes to patients receivingtreatment [21]. But these findings should be further studied. Astudy conducted on 192 orbital lymphoma patients concluded thatDLBCL have lower survival rates than MALT because they havehigher extra orbital spread rates [16]. This study also showed thatDLBCL has lower survival rates than follicular lymphoma which isinconsistent with our results. The difference in results may beexplained by the fact that we divided follicular lymphoma in grade1 and 2. Grade 1 had higher survival rates than DLBCL, whilegrade 2 had lower survival rates. It can be concluded thatlymphoma histological subtypes can be used to assess the
melanoma in age group 20–49. (B) Observed survival rates for melanoma in age
Fig. 4. Observed survival rates for different rhabdomyosarcoma subtypes for age group 0–19.
W.M. Hassan et al. / Cancer Epidemiology 38 (2014) 515–522 519
prognosis of the tumor [8,16,22]. The Kaplan Meier curves in Fig. 1can be used to predict the effect of lymphoma subtype on survivalrates. Age, sex and treatment received cannot be used as prognosticfactors in lymphoma patients [18]. However another studyproposed that muscular infiltration, ptosis and plegia can be usedas prognostic factors in lymphoma patients with various histolog-ical subtypes [23]. Histopathological staging of lymphoma is themost important step in the management of the disease. It lines upthe required steps for the proper choice of treatment modalities.Treatment can depend on local radiotherapy in low grade diseaseor may require integration between multiple modalities especiallyin moderate and high-grade disease (and when considering thepossible systemic nature of the disease). In high grade lymphomas,a consolidation using high dose chemotherapy and stem celltransplantation may be required. Treatment with targetedimmunotherapy is being investigated. Survival varies extremelyin the literature between different histological grades [24].
Table 2Relative survival rates for different histological subtypes. Comparisons between differe
Carcinomas survival
estimate (95% CI)
Melanomas su
estimate (95%
All cases 0.748 (0.686–0.8) 0.756 (0.669–
Gender
Males 0.742 (0.669–0.801) 0.742 (0.607–
Females 0.76 (0.63–0.849) 0.765 (0.641–
Race
White 0.744 (0.676–0.799) 0.752 (0.662–
Not white 0.686 (0.488–0.820) 0.731 (0.152–
Treatment
Radiation 0.564 (0.422–0.685) 0.527 (0.275–
Surgery 0.797 (0.725–0.852)b 0.8 (0.693–0.8
Surgery and radiation 0.653 (0.476–0.782) 0.315 (0.038–
a Confidence interval could not be calculated.b Statistically significant, Z values > 1.96.
Ocular surface squamous neoplasia is a group of tumors thatshows lower incidence in northern countries. It is more common inAfrica and regions with more direct frequent exposure to sunlight,and includes squamous cell carcinomas in addition to dysplasiaand carcinoma in situ [25,26]. They are associated with sunexposure and AIDS and HPV infections [27,28]. Carcinomasrepresent almost 24.2% of total cases. Literature review revealeda lack in coverage of carcinomas affecting the orbit. There was nostatistically significant difference in observed survival ratesbetween different carcinoma subtypes in the 20–49 age group.Conversely, in the older age group, the difference betweencarcinomas subtypes in observed survival rates was found to bestatistically significant (p value <0.001). In these tumors squamouscell carcinoma was the most predominant type (63.7% of totalcarcinomas patients in older age group). The statistically signifi-cant difference was probably due to higher survival rates ofsquamous cell carcinomas when compared with adenocarcinomas
nt gender, race and treatment are shown.
rvival
CI)
Rhabdomyosarcomas
survival estimate (95% CI)
Lymphomas survival
estimate (95% CI)
0.824) 0.931 (0.797–0.977) 0.907 (0.877–0.93)
0.837) 0.894 (0.703–0.965) 0.91 (0.858–0.944)
0.851) 1 (a) 0.905 (0.865–0.934)
0.821) 0.939 (0.774–0.98.5) 0.910 (0.866–0.927)
0.949) 0.900 (0.472–0.985) 0.926 (0.856–0.963)
0.728) 0.944 (0.789–0.986) 0.957 (0.913–0.978)
73)b 0.938 (0.628–0.991) 0.925 (0.868–0.958)
0.665) 0.924 (0.562–0.989) 0.958 (0.865–0.987)
W.M. Hassan et al. / Cancer Epidemiology 38 (2014) 515–522520
and adenoid cyst carcinomas. Thus carcinoma subtypes can beused to assess the prognosis of the tumor in the older age group(Fig. 2). The five-year observed survival rate was calculated for allcarcinoma patients in order to be able to compare it to rates insimilar studies. The five-year observed survival rate for allcarcinoma patients in our study was 60.4% (95% CI: 55.9–64.9).Survival rate for carcinoma subtypes included in our study atother sites in the SEER database was calculated. This resultincludes all topographies and same histological subtypes, agegroups and time period as the patients of our study. Comparisonbetween orbital carcinomas and carcinomas at other sitesrevealed that orbital carcinomas observed survival was signifi-cantly higher than carcinomas at other sites 48.6% (95% CI: 48.5–48.7). The last study describing survival rates for carcinomas inorbital sites collected data regarding 13 patients at MoorfieldsEye Hospital, United Kingdom. They stated that the five-yearsurvival rate was 72% and concluded that it has improved whencompared to previous studies (27%) due to changes in modernmanagement [3]. Results of this study are inconsistent with ourstudy as we reported a five-year survival of 60.4%. Unfortunately,direct comparisons between both results could not be done as theother study did not report confidence intervals. An earlier studycarried in 1975 that collected data from 28 patients revealed thatthe five-year survival was <20% [29]. It can be concluded thatsurvival rates of carcinoma patients has increased but not to theextent mentioned before. Surgical excision is the main line oftreatment. Due to the high recurrence rate, it is usually followedby cryotherapy, brachytherapy or topical mitomycin C [26,30–32]. Other chemotherapy options were used for decreasingrecurrence rate [33]. Although it has a high local recurrence, it israre to metastasize or cause death [34]. It was noticed in ourstudy that patients with ocular carcinoma treated by surgeryhave higher relative survival than patients treated withradiotherapy. However, this finding should be cautiouslyinterpreted since similar registry based studies cannot replaceproperly designed and implemented clinical trials.
Conjunctival melanoma is a rare entity and characteristicallyaffects older age groups [35,36]. It originates mainly fromprimary acquired melanosis [37]. In our study, the 20–49 agegroup had a five-year observed survival rate of 81.8% (95% CI:70.4–93.2), while the �50 age group had five-year observedsurvival rate of 60% (95% CI: 52.1–67.8). A study conducted in theperiod between 1969 and 1991 in Sweden showed actuarial 10-year survival proportion using life-table analysis equal to 70%[36]. Previously reported results showed up to 20% mortality atthe five-year point, which doubled at 10 years [38–40]. Surgicalexcision with proper a safety margin is the primary line ofmanagement supported by other modalities [41]. It was noticedin our study that melanoma patients treated by surgery hadhigher relative survivals than patients treated by radiotherapy.Again, this finding should be cautiously interpreted for the samereason that similar registry based studies cannot replaceproperly designed and implemented clinical trials. Local recur-rence is observed in more than half of the patients [38].Involvement of lymph nodes diagnosed by lymphoscintigraphycarries bad prognosis [42]. Systemic chemotherapy is the firstline of management when metastasis is proven.
Rhabdomyosarcomas (RMS) are found to be the predominanthistology affecting orbit in children. They can be divided intotwo main subtypes: embryonal and alveolar RMS. EmbryonalRMS are much more common than alveolar type [43,44]. Thesetumors are considered to have good prognosis. In the latestreview about orbital RMS in 2013 it stated that survival ratesincreased from 30% in 1960 to 90% currently [45]. Before the eraof radiotherapy in 1970s, exenteration was the method of choicefor the management of rhabdomyosarcoma with a 30% overall
survival [46]. Enhancements of radiotherapy have saved lot oflives and eyes. Efforts were paid by Intergroup Rhabdomyosar-coma Study Group to improve classification and staging andhence proper outcome prediction and choice of treatment [44].Introduction of chemotherapy increased survival to 90% in somegroups [43]. Current studies are focusing on improving thetreatment and decreasing long-term side effects [47,48].Embryonal RMS had higher a five-year observed survival rateof 90.9% (95% CI: 82.3–99.5) than the alveolar type (87.5% (95%CI: 64.6–100*)). This difference was statistically insignificant(Fig. 4). These results are inconsistent with similar studies whichstated that there was a statistically significant differencebetween survivals in patients suffering from both histologicalsubtypes. Inconsistency of our results with similar studies isprobably due to the small number of those patients in our study(N = 8). Alveolar RMS had an average of 74% five-year observedsurvival rate in other studies, and they were statisticallydifferent from embryonal RMS who had an average five-yearsurvival rate of 94%. This gave to the conclusion that embryonalRMS histology can be used as a favorable prognostic factor [43–45]. Several other factors were identified as prognostic factorsdeterminants in other studies. Extension of disease and tumorsize were identified as factors in some studies [44,49,50].Another study concluded that an age less than 3 years old was aless favorable prognostic factor [43].
5. Conclusion
At orbital and lacrimal gland sites in adults, lymphomas havebetter survival rate than carcinomas. Lymphoma subtypes can beused as a determinant prognostic factor, while carcinoma subtypescan only be used as a determinant prognostic factor in older agegroups.
In children, rhabdomyosarcomas have generally high survivalrates. Embryonal rhabdomyosarcoma has better survival rate thanalveolar rhabdomyosarcoma.
Conflict of interest of statement
None declared.
Authorship contribution
WMH and ASA: study concept and design; WMH and ASA:review of literature, and drafting of the manuscript; WMH andMSB: statistical analysis; ASA and SE: critical revision of themanuscript for important intellectual content; WMH, ASA, MSBand SE: final approval of the version to be published; WMH andASA equally participated to this paper.
References
[1] Volpe NJ, Jakobiec FA. Pediatric orbital tumors. Int Ophthalmol Clin1992;32(January (1)):201–21 [Internet]. Available from: http://www.pubmed-central.nih.gov/articlerender.fcgi?artid=3610355&tool=pmcentrez&render-type=Abstract.
[2] Johnson TE, Senft SH, Nasr AM, Bergqvist G, Cavender JC. Pediatric orbitaltumors in Saudi Arabia. Orbit 1990;9(January (4)):205–15 [Internet]. Availablefrom: http://informahealthcare.com/doi/abs/10.3109/01676839009019286.
[3] Mehta JS, Abou-Rayyah Y, Rose GE. Orbital carcinoid metastases. Ophthalmol-ogy 2006;113(March (3)):466–72 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16458966 [cited 30.09.13].
[4] Parkin DM, Hakulinen T. Cancer registration: principles and methods. IARCScientific Publications No. 95. Lyon: International Agency for Research onCancer, 1991. 159–76.
[5] Database: Incidence – SEER 18 Regs Research Data, November 2011 Sub(1973–2010) <Katrina/Rita Population Adjustment> – Linked To CountyAttributes – Total U.S. 1969–2010. Surveillance, Epidemiology, and End
W.M. Hassan et al. / Cancer Epidemiology 38 (2014) 515–522 521
Results (SEER) Program SEER*Stat; 2013 [Internet]. Available from:www.seer.cancer.gov.
[6] Site S, List HV, List V, Program S, Control Q, Suite S, et al. ICD-0-3 SEER site/histology validation list; 2012;301–496, December 5.
[7] Cho EY, Han JJ, Ree HJ, Ko YH, Kang Y, Ahn HS, et al. Clinicopathologic analysisof ocular adnexal lymphomas: extranodal marginal zone b-cell lymphomaconstitutes the vast majority of ocular lymphomas among Koreans andaffects younger patients. Am J Hematol 2003;73(June (2)):87–96 [Internet].Available from: http://onlinelibrary.wiley.com/doi/10.1002/ajh.10332/pdf[cited 24.09.13].
[8] Coupland SE, Krause L, Delecluse HJ, Anagnostopoulos I, Foss HD, Hummel M,et al. Lymphoproliferative lesions of the ocular adnexa. Analysis of 112 cases.Ophthalmology 1998;1430–41 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/9709754.
[9] Moslehi R, Coles FB, Schymura MJ. Descriptive epidemiology of ophthalmicand ocular adnexal non-Hodgkin’s lymphoma. Expert Rev Ophthalmol2011;6(April (2)):175–80 [Internet]. Available from: http://www.expert-reviews.com/doi/abs/10.1586/eop.11.16.
[10] Smedby KE, Hjalgrim H. Epidemiology and etiology of mantle cell lymphomaand other non-Hodgkin lymphoma subtypes. Semin Cancer Biol 2011;21(No-vember (5)):293–8 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21945518 [cited 30.09.13].
[11] Bolek TW, Moyses HM, Marcus RB, Gorden L, Maiese RL, Almasri NM, et al.Radiotherapy in the management of orbital lymphoma. Int J Radiat Oncol BiolPhys 1999;44(April (1)):31–6 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/10219791.
[12] White WL, Ferry JA, Harris NL, Grove AS. Ocular adnexal lymphoma.A clinicopathologic study with identification of lymphomas of mucosa-associated lymphoid tissue type. Ophthalmology 1995;102(December(12)):1994–2006 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/9098307 [cited 30.09.13].
[13] Auw-Haedrich C. Long term outcome of ocular adnexal lymphoma subtypedaccording to the REAL classification. Br J Ophthalmol 2001;85(January (1)):63–69 [Internet]. Available from: http://bjo.bmj.com/cgi/doi/10.1136/bjo.85.1.63 [cited 30.09.13].
[14] Cahill M, Barnes C, Moriarty P, Daly P, Kennedy S. Ocular adnexal lymphoma –comparison of MALT lymphoma with other histological types. Br J Ophthalmol1999;83(June (6)):742–7 [Internet]. Available from: http://bjo.bmj.com/cgi/doi/10.1136/bjo.83.6.742 [cited 25.09.13].
[15] Alkatan HM, Alaraj AM, Al-Ayoubi A. Diffuse large B-cell lymphoma of theorbit: a tertiary eye care center experience in Saudi Arabia. Saudi J Ophthalmol2002;26(April (2)):235–9 [Internet]. King Saud University; Available from:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3729852&tool=pmcentrez&rendertype=Abstract [cited 30.09.13].
[16] Jenkins C. Histological features of ocular adnexal lymphoma (REAL classifica-tion) and their association with patient morbidity and survival. Br J Ophthal-mol 2000;84(August (8)):907–13 [Internet]. Available from: http://bjo.bmj.com/cgi/doi/10.1136/bjo.84.8.907 [cited 25.09.13].
[17] Fung CY, Tarbell NJ, Lucarelli MJ, Goldberg SI, Linggood RM, Harris NL, et al.Ocular adnexal lymphoma: Clinical behavior of distinct World Health Orga-nization classification subtypes. Int J Radiat Oncol Biol Phys 2003;57(Decem-ber (5)):1382–91 [Internet]. Available from: http://linkinghub.elsevier.com/retrieve/pii/S0360301603007673 [cited 30.09.13].
[18] Nola M, Lukenda A, Bollmann M, Kalauz M, Petrovecki M, Bollmann R.Outcome and prognostic factors in ocular adnexal lymphoma. Croatian MedJ 2004;45(June (3)):328–32 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15185428.
[19] Liao Z, Ha CS, McLaughlin P, Manning JT, Hess M, Cabanillas F, et al. Mucosa-associated lymphoid tissue lymphoma with initial supradiaphragmatic pre-sentation: natural history and patterns of disease progression. Int J RadiatOncol Biol Phys 2000;48(September (2)):399–403 [Internet]. Available from:http://www.ncbi.nlm.nih.gov/pubmed/10974453.
[20] Rootman DB, Mavrikakis I, Connors JM, Rootman J. Primary, unilateral ocularadnexal lymphoma: disease progression and long-term survival. OphthalmicPlast Reconstr Surg 2011;27(6):405–9 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21659919 [cited 30.09.13].
[21] Tanimoto K, Kaneko a, Suzuki S, Sekiguchi N, Maruyama D, Kim SW, et al.Long-term follow-up results of no initial therapy for ocular adnexalMALT lymphoma. Ann Oncol 2006;17(January (1)):135–40 [Internet].Available from: http://www.ncbi.nlm.nih.gov/pubmed/16236754 [cited30.09.13].
[22] Jenkins C, Rose GE, Bunce C, Cree I, Norton a, Plowman PN, et al. Clinicalfeatures associated with survival of patients with lymphoma ofthe ocular adnexa. Eye (Lond) 2003;17(October (7)):809–20 [Internet].Available from: http://www.ncbi.nlm.nih.gov/pubmed/14528242 [cited25.09.13].
[23] Martinet S, Ozsahin M, Belkacemi Y, Landmann C, Poortmans P, Oehlere C, et al.Outcome and prognostic factors in orbital lymphoma. Int J Radiat Oncol BiolPhys 2003;55(March (4)):892–8 [Internet]. Available from: http://linkinghu-b.elsevier.com/retrieve/pii/S0360301602041597 [cited 25.09.13].
[24] Stacy RC, Jakobiec FA, Herwig MC, Schoenfield L, Singh A, Grossniklaus HE.Diffuse large B-cell lymphoma of the orbit: clinicopathologic, immunohisto-chemical, and prognostic features of 20 cases. Am J Ophthalmol 2012;154(July(1)):87–980 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/22503694 [cited 19.04.14].
[25] Lee GA, Hirst LW. Ocular surface squamous neoplasia. Surv Ophthalmol1995;39(6):429–50 [Internet]. Available from: http://www.ncbi.nlm.nih.-gov/pubmed/7660300 [cited 19.04.14].
[26] Pe’er J. Ocular surface squamous neoplasia. Ophthalmol Clin N Am2005;18(March (1)):1–13. vii, [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15763187 [cited 19.04.14].
[27] McDonnell JM, Mayr AJ, Martin WJ. DNA of human papillomavirus type 16 indysplastic and malignant lesions of the conjunctiva and cornea. N Engl J Med1989;320(June (22)):1442–6 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/2541337 [cited 19.04.14].
[28] Goedert JJ, Cote TR. Conjunctival malignant disease with AIDS in USA. Lancet1995;346(July (8969)):257–8 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/7616836 [cited 19.04.14].
[29] Ferry aP, Font RL. Carcinoma metastatic to the orbit. Mod Probl Ophthalmol1975;14(January):377–81 [Internet]. Available from: http://www.ncbi.nlm.-nih.gov/pubmed/953971.
[30] Fraunfelder FT, Wallace TR, Farris HE, Watkins J, Hendrickson R, Smead WJ,et al. The role of cryosurgery in external ocular and periocular disease. TransSect Ophthalmol Am Acad Ophthalmol Otolaryngol 1977;83(4 Pt 1):713–24[Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/898492[cited 19.04.14].
[31] Zehetmayer M, Menapace R, Kulnig W. Combined local excision and brachy-therapy with ruthenium-106 in the treatment of epibulbar malignancies.Ophthalmologica 1993;207(January (3)):133–9 [Internet]. Available from:http://www.ncbi.nlm.nih.gov/pubmed/8278179 [cited 19.04.14].
[32] Frucht-Pery J, Rozenman Y, Pe’er J. Topical mitomycin-C for partially excisedconjunctival squamous cell carcinoma. Ophthalmology 2002;109(March(3)):548–52 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/11874760 [cited 19.04.14].
[33] Kim JW, Abramson DH. Topical treatment options for conjunctival neoplasms.Clin Ophthalmol 2008;2(September (3)):503–15 [Internet]. Available from:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2694026&-tool=pmcentrez&rendertype=Abstract [cited 19.04.14].
[34] Iliff WJ, Marback R, Green WR. Invasive squamous cell carcinoma of theconjunctiva. Arch Ophthalmol 1975;93(February (2)):119–22 [Internet]. Avail-able from: http://www.ncbi.nlm.nih.gov/pubmed/1115670 [cited 19.04.14].
[35] De Wolff-Rouendaal D. Conjunctival melanoma in The Netherlands: aclinico-pathological and follow-up study. Orbit 1990;9(January (3)):177-177 [Internet]. Available from: http://informahealthcare.com/doi/abs/10.3109/01676839009012366 [cited 19.04.14].
[36] Seregard S, Kock E. Conjunctival malignant melanoma in Sweden 1969–91.Acta Ophthalmol 2009;70(May (3)):289–96 [Internet]. Available from: http://doi.wiley.com/10.1111/j. 1755-3768.1992.tb08566.x [cited 19.04.14].
[37] Shields CL, Markowitz JS, Belinsky I, Schwartzstein H, George NS, Lally SE,et al. Conjunctival melanoma: outcomes based on tumor origin in 382consecutive cases. Ophthalmology 2011;118(February (2)):389–95. e1-2, [In-ternet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20723990 [cit-ed 19.04.14].
[38] Seregard S. Conjunctival melanoma. Surv Ophthalmol 1998;42(4):321–50[Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/9493274[cited 19.04.14].
[39] Shields CL. Conjunctival melanoma: risk factors for recurrence, exenteration,metastasis, and death in 150 consecutive patients. Trans Am Ophthalmol Soc2000;98(January):471–92 [Internet]. Available from: http://www.pubmed-central.nih.gov/articlerender.fcgi?artid=1298238&tool=pmcentrez&renderty-pe=Abstract [cited 19.04.14].
[40] Paridaens AD, Minassian DC, McCartney AC, Hungerford JL. Prognosticfactors in primary malignant melanoma of the conjunctiva: a clinicopatho-logical study of 256 cases. Br J Ophthalmol 1994;78(April (4)):252–9[Internet]. Available from: http://www.pubmedcentral.nih.gov/articleren-der.fcgi?artid=504757&tool=pmcentrez&rendertype=Abstract [cited19.04.14].
[41] Shields JA, Shields CL, De Potter P. Surgical management of conjunctivaltumors. The 1994 Lynn B. McMahan lecture. Arch Ophthalmol 1997;115(June(6)):808–15 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/9194740 [cited 19.04.14].
[42] Tatla T, Hungerford J, Plowman N, Ghufoor K, Keene M. Conjunctivalmelanoma: the role of conservative surgery and radiotherapy in regionalmetastatic disease. Laryngoscope 2005;115(May (5)):817–22 [Internet].Available from: http://www.ncbi.nlm.nih.gov/pubmed/15867646 [cited19.04.14].
[43] Oberlin O, Rey a, Anderson J, Carli M, Raney RB, Treuner J, et al. Treatment oforbital rhabdomyosarcoma: survival and late effects of treatment – resultsof an international workshop. J Clin Oncol 2001;19(January (1)):197–204 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/11134213.
[44] Kodet R, Newton WA, Hamoudi AB, Asmar L, Wharam MD, Maurer HM. Orbitalrhabdomyosarcomas and related tumors in childhood: relationship of mor-phology to prognosis – an Intergroup Rhabdomyosarcoma Study. Med PediatrOncol 1997;29:51–60 [Internet]. Available from: http://www.ncbi.nlm.nih.-gov/pubmed/9142207.
[45] Jurdy L, Merks JHM, Pieters BR, Mourits MP, Kloos RJHM, Strackee SD, et al.Orbital rhabdomyosarcomas: a review. Saudi J Ophthalmol 2013;27(July(3)):167–75 [Internet]. Available from: http://linkinghub.elsevier.com/re-trieve/pii/S1319453413000544 [cited 30.09.13].
W.M. Hassan et al. / Cancer Epidemiology 38 (2014) 515–522522
[46] Knowles DM, Jakobiec FA, Potter GD, Jones IS. Ophthalmic striated muscleneoplasms. Surv Ophthal 1976;21(3):219–61 [Internet]. Available from:http://www.ncbi.nlm.nih.gov/pubmed/797029 [cited 19.04.14].
[47] Wolden SL, Wexler LH, Kraus DH, Laquaglia MP, Lis E, Meyers PA. Intensity-modulated radiotherapy for head-and-neck rhabdomyosarcoma. Int J RadiatOncol Biol Phys 2005;61(April (5)):1432–8 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15817347 [cited 19.04.14].
[48] Yock T, Schneider R, Friedmann A, Adams J, Fullerton B, Tarbell N. Protonradiotherapy for orbital rhabdomyosarcoma: clinical outcome and a dosimet-ric comparison with photons. Int J Radiat Oncol Biol Phys 2005;63(November
(4)):1161–8 [Internet]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15950401 [cited 19.04.14].
[49] Lawrence W, Gehan EA, Hays DM, Beltangady M, Maurer HM. Prognosticsignificance of staging factors of the UICC staging system in childhoodrhabdomyosarcoma: a report from the Intergroup Rhabdomyosarcoma Study(IRS-II). J Clin Oncol 1987;5(January (1)):46–54 [Internet]. Available from:http://www.ncbi.nlm.nih.gov/pubmed/3543238 [cited 28.09.13].
[50] Crist BWM, Garnsey L, Beltangady MS, Gehan E, Ruymann F, Webber B, et al.Prognosis in children with rhabdomyosarcoma: a report of the intergrouprhabdomyosarcoma studies I and II. J Clin Oncol 2013;8(3):443–52.