doi:10.1016/S0360-3016(03)00773-9

CLINICAL INVESTIGATION Head and Neck

NASOPHARYNGEAL CARCINOMA IN CHILDREN: TEN YEARS’EXPERIENCE AT THE TATA MEMORIAL HOSPITAL, MUMBAI

SIDDHARTHA LASKAR, M.D. (R.T.),* V IMAL SANGHAVI , M.D. (R.T.),*MARRY ANN MUCKADEN, M.D. (R.T.),* SARBANI GHOSH (LASKAR), M.D. (R.T.),*

VISHAL BHALLA , M.D. (R.T.),* SHRIPAD BANAVALI , M.D. (PEDIATRICS),†

PURNA KURKURE, M.D. (PEDIATRICS),† CHANDRIKA NAIR, D.C.H.,† AND

KETAYUN A. DINSHAW, F.R.C.R.*

Departments of *Radiation Oncology and†Medical Oncology, Tata Memorial Hospital, Mumbai, India

Purpose: To evaluate the disease characteristics and outcome of children with nasopharyngeal carcinoma treatedat the Tata Memorial Hospital, Mumbai.Methods and Materials: Between 1990 and 2000, 81 pediatric patients with a diagnosis of nasopharyngealcarcinoma were treated at the Tata Memorial Hospital. The median age was 14 years. The male/female ratio was2.8:1. Of the 81 patients, 32 (39%), 21 (26%), and 28 (35%) had T1-T2, T3, and T4 (TNM International UnionAgainst Cancer staging system, 1997), respectively. Ninety-one percent presented with nodal metastasis. Thirtypatients (37%) had lymph nodes>6 cm, and 45 (56%) had bilateral nodes at presentation. Histologically, 77patients (95%) had undifferentiated carcinoma. Eighty-five percent received neoadjuvant multiagent chemo-therapy containing bleomycin, methotrexate, and cisplatin, followed by radiotherapy (RT).Results: After a median follow-up of 50 months, the disease-free survival (DFS) and overall survival (OS) ratefor the entire group was 45% and 54%, respectively. Kaplan-Meier curves were used for evaluation of prognosticfactors and were compared using the log–rank test. Nodal status had a significant impact on DFS (p � 0.021) andOS (p � 0.006). Complete responders to chemotherapy had superior DFS (p � 0.000) and OS (p � 0.000). RTdoses>60 Gy resulted in better DFS (p � 0.020) and OS (p � 0.012). Combined chemotherapy plus RT resultedin improved DFS (p � 0.457) and OS (p � 0.296), although the difference was not statistically significant.Conclusion: Combined modality management using chemotherapy and RT resulted in satisfactory locoregionalcontrol and OS in pediatric patients with nasopharyngeal carcinoma. Nodal involvement, response to chemo-therapy, and RT dose were important prognostic factors. © 2004 Elsevier Inc.

Nasopharyngeal carcinoma, Childhood, Combined modality.

INTRODUCTION

Nasopharyngeal cancers account for 1% of all childhoodcancers and 40–50% of all malignancies involving thenasopharynx in children (1–3). The incidence of nasopha-ryngeal carcinoma (NPC) in the pediatric age group variesgreatly according to racial and geographic factors. Althoughmost patients with NPC present in the fourth to fifth decade,the classic age incidence graph is bimodal, with an initialpeak in late childhood (1, 4–7). The most common histo-logic type seen in children is undifferentiated carcinoma(World Health Organization Type III), which is associatedwith advanced locoregional disease at presentation and agreater rate of distant metastasis (2, 3, 8). Unlike the well-differentiated squamous carcinoma variant seen in most

adults, these tumors have a greater association with priorEpstein-Barr virus infection (9, 10).

For many years, external beam radiotherapy (RT) hasbeen the mainstay of treatment for patients with NPC.Despite the high doses of radiation that frequently result insignificant late morbidity in long-term survivors (2, 3, 8), alarge proportion of patients with advanced disease havelocoregional failure with or without distant failure (2, 3, 8).Standard therapy for NPC in children has generally fol-lowed the guidelines established for adults. The poor overallsurvival (OS) rates have prompted investigators to evaluatethe role of chemotherapy before (11), concurrent with (12,13), and after RT (14).

Because of the rarity of the disease, published clinical

Reprint requests to: Siddhartha Laskar, M.D. (R.T.), Depart-ment of Radiation Oncology, Tata Memorial Hospital, Dr ErnestBorges Marg Rd., Parel, Mumbai 400 012 India. Tel: 091 02224177167; Fax: 091 022 4146937; E-mail: laskars2000@yahoo.com

Received Apr 10, 2003, and in revised form Jun 9, 2003.Accepted for publication Jun 13, 2003.

Int. J. Radiation Oncology Biol. Phys., Vol. 58, No. 1, pp. 189–195, 2004Copyright © 2004 Elsevier Inc.

Printed in the USA. All rights reserved0360-3016/04/$–see front matter

189

trials in this age group are scanty. Most published studiesconsisted of retrospective reviews of a small number ofpatients, treated without uniform regimens, without addi-tional chemotherapy, and of patients accrued over longperiods (15–19). The impact of chemotherapy on long-termdisease-free survival (DFS) remains to be answered bywell-conducted randomized trials. To date, no randomizedclinical trial has been published addressing these issues inchildren. However, as is the case with other rare malignan-cies, single-institutional reports could contribute to our un-derstanding of the natural biology of the disease and newtreatment strategies that could provide a scientific basis foradditional multicenter studies.

In this paper, we reviewed 10 years of experience oftreating children with nasopharyngeal cancer at the TataMemorial Center, Mumbai, using a uniform treatment reg-imen. We reviewed in detail the epidemiologic, clinical, andtreatment-related aspects in our patients.

METHODS AND MATERIALS

PatientsBetween January 1990 and May 2000, 81 children �18

years old diagnosed with NPC were treated at the TataMemorial Hospital, Mumbai. Of the 81 patients, 60 wereboys and 21 were girls, a male/female ratio of 2.8:1. The agerange was 8–18 years (median 15). Most patients were fromthe state of Maharashtra and the eastern part of India. Theduration of symptoms at presentation ranged from 1 to 12months (median 4).

The extent of disease was determined by a completeclinical examination, including indirect mirror examinationof nasopharynx. Complete hematologic evaluation andblood chemistry were routinely assessed. A plain film X-rayof the nasopharynx was taken for all patients. These werecomplemented with CT of the nasopharynx, base of theskull, and neck in 40 patients (49%) who could afford it.Chest radiography and bone marrow biopsy were done aspart of the metastatic workup. The histologic diagnosis wasconfirmed in all cases from biopsy of the primary lesion ormetastatic lymph nodes. The TNM 1997 classification sys-tem of the American Joint Committee on Cancer (AJCC)staging and End Results Reporting (20) was used to definethe disease extent.

The treatment protocol included a combination of twocycles of neoadjuvant chemotherapy followed by RT formost children; 8 (10%) received RT alone.

The chemotherapy regimen consisted of a three-drugcombination of i.v. cisplatin (20 mg/m2 on Days 1–5), i.v.bleomycin (10 U on Days 1 and 8), and i.v. methotrexate(50 mg/m2 on Days 1 and 8) with premedication and hy-dration. Each cycle was repeated after 4 weeks. RT wasdelivered to the nasopharynx and lymphatic drainage areasusing 60Co �-rays or 6-MV photons with bilateral parallel-opposed fields. In patients in whom it was not possible tocover the lower neck adequately with bilateral ports, thelower cervical and supraclavicular regions were treated us-

ing a single anterior port with a midline shield for the larynxand spinal cord. In patients with clinical or radiologic (CTor plain radiography) features suggestive of intracranialextension, the superior border was extended superiorly toinclude the disease extension in the RT portal adequately.Tissue compensators were used in the lateral ports. A thirdanterior field was used in patients with tumor extension intothe nasal cavity or ethmoid sinuses. Patients were immobi-lized using plaster of Paris casts or thermoplastic molds.The target volume of RT included the entire nasopharynx,posterior part of the nasal cavity (whole nasal cavity whennasal extension was present), postethmoid, sphenoid, basi-occiput, base of skull, pterygoid fossa, posterior third of themaxillary sinuses (whole sinus when extension present),lateral and posterior pharyngeal wall to the level of themid-tonsillar fossa, retropharyngeal lymph nodes, and cer-vical group of neck nodes, except for the submental node. Atotal dose of 55–70 Gy was administered in daily fractionsof 1.8–2 Gy, 5 d/wk. The spinal cord was shielded after 46Gy, and the posterior lymphatic chains were treated withelectrons (6–12 million electron volt [Mev]). Clinicallyinvolved neck areas received 60–66 Gy.

After completion of the planned therapy, patients werefollowed every 3 months for the first year, every 6 monthsfor the next 2 years, and yearly thereafter. Response wasdocumented according to the World Health Organizationresponse assessment criteria (21). A complete response wasdefined as the disappearance of all evident disease. A de-crease of �50% was defined as a partial response, anddecrease of �50% was considered no response/stable dis-ease. A progressive increase in the size of the tumor or theappearance of new lesions was defined as progressive dis-ease. The Kaplan-Meier method (22) was used to calculatethe actuarial survival. The persistence of disease and theoccurrence of relapses were the end points for DFS, anddeath from any cause was considered an event for thecalculation of OS. The differences between the curves werecompared using the log–rank test (23). The survival analysiswas done using the Statistical Package for Social Sciences,version 10.0.

RESULTS

Eighty-one patients who were �18 years of age ac-counted for 11% of all cases of NPC and 1.5% of allpediatric malignancies seen at the Tata Memorial Hospitalannually (24).

Pain was the main symptom at presentation in 72 children(89%), presenting as headache, otalgia, or pain in metastaticlymph nodes. Nasal symptoms (obstruction, epistaxis) werepresent in 55%. Sixty-eight children (84%) presented withthe complaint of swelling in the neck. Fever and weight losswere present in 42%. In addition 2 patients (2.4%) hadcranial nerve palsy at presentation. The median duration ofsymptoms at presentation was 4 months.

Nasopharynx was the only site of involvement in 12patients (15%), and bone and paranasal sinuses were in-

190 I. J. Radiation Oncology ● Biology ● Physics Volume 58, Number 1, 2004

volved in 14 (17%) and 17 (21%), respectively. Intracranialextension was seen in 11 patients (14%), and involvementof the infratemporal fossa and extension to the orbit wasseen in 1 patient each. Seventy-four patients (91%) hadclinically palpable nodes at presentation, and 45 patients(56%) had bilateral neck node involvement. Thirty patients(37%) had neck nodes measuring �6 cm in the greatestdimension. Thirty-seven patients (46%) had neck nodeswith restricted mobility or fixed to underlying structures.Four patients (5%) presented with distant metastasis, ofwhom 1 patient had bone metastasis only, 1 patient each hadlung and liver metastasis in addition to bone metastasis, and1 patient presented with widely disseminated disease withinvolvement of multiple bones, lung, liver, ascites, andmultiple s.c. metastases. The bone marrow was involved atpresentation in 2 patients (2.4%).

On the basis of the disease extent at presentation, patientswere grouped according to the American Joint Commission onCancer TNM classification system. The stage distribution isshown in Table 1. Thirty-two patients (39%) had T1-T2 le-sions, 21 (26%) had T3 lesions, and 28 (35%) had T4 disease.Most patients presented with Stage III-IV disease.

Of the 81 patients, 64 (79%) had adequate data availablefor survival analysis, and 57 (89%) of whom had received acombination of chemotherapy and RT and 7 (11%) hadreceived RT alone.

Of the patients who received neoadjuvant chemotherapy,24 (42%) achieved a complete response, 25 (44%) a partialresponse, and 8 (14%) had no response or progressivedisease after completion of two cycles of chemotherapy.After completion of the planned treatment, 55 patients(68%) achieved a complete response, 6 (7%) a partial re-sponse, and 3 (4%) had no response or progressive disease.

After a median follow-up of 50 months for the survivingpatients, the DFS and OS rate at 5 years was 45% and 54%,respectively. At last follow-up, 33 patients (52%) were free ofdisease. Of the 31 patients with failure, 12 (39%) had localfailure, 16 (52%) had nodal failure, and 17 (55%) had distantfailure. The pattern of failure is summarized in Table 2. Uni-variate analysis was done to evaluate the various prognosticfactors for their impact on DFS and OS (Table 3).

The prognostic factors that had a statistically significantinfluence on DFS were patient age at presentation (p � 0.031),

nodal stage (p � 0.021), TNM stage (p � 0.020), size ofinvolved nodes (p � 0.004), nodal fixity (p � 0.002), RT dose(p � 0.020), and response to chemotherapy (p � 0.000; Fig.1a). Gender, T stage, bilateral nodal involvement, and treat-ment modality did not have a significant impact on DFS (Fig.

Table 1. AJCC TNM stage distribution

T stage

N stage (n)

N0 N1 N2 N3Total(%)

T1-T2 2 5 11 14 32 (39)T3 2 3 11 5 21 (26)T4 3 4 10 11 28 (35)Total (%) 7 (9) 12 (15) 32 (39) 30 (37) 81 (100)

Stage I-II, 7/81 (9%); Stage III, 27/81 (33%); Stage IV, 47/81(58%).

Abbreviation: AJCC � American Joint Commission on Cancer.

Table 2. Failure patterns

Failure n (%)

None 33 (52)Local 3 (10)Nodal 7 (23)Local � nodal 4 (13)Distant 12 (39)Local � nodal � distant 5 (16)Total with failure 31 (100)

Table 3. Prognostic factors

DFS (%) p OS (%) p

Age (y)�12 78 0.030 77 0.378�12 54 54

GenderMale 39 0.062 53 0.045Female 63 78

T stageT1–T2 55

0.38772

0.320T3 53 66T4 27 42

N stageN0 80

0.021

80

0.006N1 64 66N2 53 55N3 22 24

Stage groupI–II 80

0.02080

0.010III 62 64IV 27 40

N size (cm)�6 55 0.004 75 0.007�6 22 71

BilateralityUnilateral 53 0.091 71 0.045Bilateral 37 52

Node fixityMobile 57 0.002 71 0.009Tethered/fixed 23 52

TreatmentRT only 28 0.457 36 0.296CT � RT 49 67

RT dose (Gy)�60 68 0.020 73 0.012�60 40 46

Response to CHTCR 69

0.00086

0.000PR 36 61NR 25 16

Abbreviations: DFS � disease-free survival; OS � overallsurvival; RT � radiotherapy; CHT � chemotherapy; CR � com-plete response; PR � partial response; NR � no response.

191NPC in childhood ● S. LASKAR et al.

2). Apart from the patient’s age at presentation, tumor stage,and treatment modality, all other prognostic factors had astatistically significant influence on OS.

During chemotherapy, the major adverse effect docu-mented was Grade 1 and 2 neutropenia (21%). The acuteeffects during RT were primarily associated with the skinand mucosa. Sixty-three percent had Grade 2 skin toxicity,and 59% had Grade 22 mucositis during or at RT comple-tion. The documented late effects included xerostomia, s.c.fibrosis, and dental carries. The treatment-related acute andlate effects are presented in Table 4.

DISCUSSION

Nasopharyngeal cancer is a relatively rare childhood ma-lignancy. In China, where NPC is one of the most common

cancers, the incidence of childhood NPC represents onlyabout 2% of all cases of NPC. In the United States andUnited Kingdom, pediatric patients comprise 10% and2.4%, respectively, of all NPC cases (1, 25). At the TataMemorial Hospital, India, NPC in children �18 years ofage accounts for 11% of cases of NPC. The age incidencegraph of patients with NPC registered at the Tata MemorialHospital showed a bimodal age distribution with two peaks,respectively, during the second and fifth decades. This isunlike that reported from China (26). We recorded a male/female ratio of 2.8:1. This ratio could partly be accountedfor by the preferential referral of males to hospitals by theilliterate and underprivileged communities because of socialbeliefs and customs.

The histopathologic type was confirmed in our institute for

Fig. 1. Kaplan-Meier curves showing comparison of (a) DFS and (b) OS in groups achieving complete response (CR),partial response (PR), or no response (NR) after chemotherapy. DFS � disease-free survival; OS � overall survival.

192 I. J. Radiation Oncology ● Biology ● Physics Volume 58, Number 1, 2004

Fig. 2. Kaplan-Meier curves showing comparison of (a) DFS and (b) OS in groups receiving RT alone and combinedRT and chemotherapy (CT).

Table 4. Acute and late toxicity caused by chemoradiation (RTOG)

Grade 1 (%) Grade 2 (%) Grade 3 (%) Grade 4 (%) NK (%)

Acute toxicityNeutropenia 21%* 21%* — — 15/64 (23)Skin 3/64 (6) 40/64 (63) 8/64 (13) — 4/64 (6)Mucosa 9/64 (14) 38/64 (59) 5/64 (8) — 12/64 (19)

Late toxicitySubcutaneous fibrosis 35/64 (55) 15/64 (23) — — —Xerostomia 54/64 (84) 10/64 (16) — — —Dental caries 10/64 (16) 8/64 (13) — — —

* Grade 1 and 2 combined.Abbreviation: RTOG � Radiation Therapy Oncology Group.

193NPC in childhood ● S. LASKAR et al.

all patients before the onset of any treatment. Ninety-onepercent of the children had undifferentiated carcinoma (WorldHealth Organization Type III). Despite not being able to do CTevaluations in all patients, most of our patients were grouped inStage III and IV on the basis of the clinical evaluation. Mostseries have reported a very small proportion of patients withStage I and II disease at presentation. In the current series, 12%of the children had early-stage disease at presentation. Despitea higher incidence of advanced locoregional disease in childrencompared with adults, the OS rates were not significantlydifferent between the two.

The chemotherapy regimen used in our patients issomewhat unusual considering the inclusion of bleomy-cin in the regimen. All three agents (i.e., platinum, bleo-mycin, and methotrexate) have been shown to be effec-tive agents in NPC (27). Although no trial has reportedusing the same regimen in children, we tried to extrapo-late the results of treating adults with NPC with the samemultidrug regimen.

The 5-year DFS of 46% and OS of 54% in our seriesmatches well with that of the other large series, whichreported DFS and OS rates in the range of 29 –60% and45–55%, respectively (2, 28, 29). In our series, age atdiagnosis had a significant impact on DFS and OS. Pa-tients �12 years of age fared better than those between12 and 18 years. Ingersoll et al. (2) also reported similarresults in their report of 57 patients. The TNM stage wasfound to have a significant impact on DFS and OS.Although patients with a lower T stage had better DFSand OS, the difference was not statistically significant.This could probably be attributed to the inadequate Tstaging in 41 patients (51%) who had not undergone CTfor primary staging. It could also be a reflection of theinadequate numbers of patients required to reach statis-tical significance. Nodal status was found to be one of themost important prognostic factors in our series influenc-ing DFS and OS. Patients with negative neck nodes hada statistically significant superior DFS and OS compared withpatients with N1-N3 nodes. This is similar to the reported5-year OS rate of 69% and 28% (p � 0.001) for patients withN0-N1 and N2-N3 nodes, respectively, reported by Sahraoui etal. (29). All 17 patients in our series who had distant failurehad clinically palpable neck nodes at presentation. Of the 17patients with distant failure, 10 (59%) had fixed/tetherednodes, 13 (78%) had bilateral neck nodes, and 8 (47%) hadnodes �6 cm at presentation; 12 (71%) had distant failurealone. The extent of neck nodal involvement (i.e, size, fixity,and bilaterality) was uniformly a significant indicator of poorprognosis.

Several investigators have looked at the correlationamong RT dose, local failure, and OS. The optimal RTdose is still a point of debate, particularly when combinedwith chemotherapy. Ingersoll et al. (2) and Martin andShah (30) reported improved local control with higherdoses. Ayan and Altun (28) also reported improved localcontrol with radiation doses �60 Gy. In our current

series, we found a statistically significant improvement inboth DFS and OS for patients receiving radiation doses�60 Gy compared with those receiving less. Adequatelocal control has also been reported for a wide range ofradiation doses of 35–65 Gy (4, 31–33). However, thehigher doses are of particular concern in the pediatric agegroup (2, 3, 8, 18, 34, 35).

Two large prospective randomized controlled trials havecompared combination chemotherapy and RT with RTalone (36, 37). Although both trials primarily included adultpatients, the trial by the Intergroup Nasopharyngeal CancerStudy Group included patients with undifferentiated carci-noma, which is the major histopathologic type seen inchildren (37). Both these trials reported an improvement inDFS, and the Intergroup Nasopharyngeal Cancer StudyGroup trial also showed an improvement in OS for thegroup receiving chemotherapy plus RT. In our current se-ries, we also found a trend toward improved DFS and OS inthe group receiving chemotherapy plus RT compared withthose receiving RT alone, although it did not reach statisti-cal significance, probably because of the small number ofpatients receiving RT alone.

Distant metastasis remains a major obstacle in the cure ofchildhood NPC (1, 2, 31, 38). In our series 17 (27%) of 64patients had failure at distant sites with or without locore-gional failure.

Data are scarce involving the long-term late effects inchildren with NPC. The reported late effects involve soft-tissue fibrosis, dental problems, skeletal deformities, thyroiddysfunction, endocrine disorders, and second cancers (2, 3,8, 33, 39, 40). In our series, although all patients developedsome degree of dryness of mouth, impaired taste sensation,s.c. fibrosis, and dental problems, no osteoradionecrosis ordevelopment of second malignancy was reported. We hadnot prospectively recorded endocrine effects and thyroiddysfunction in our patients.

CONCLUSION

The results of this 10-year study from a single institutionsuggest that local RT combined with multiagent chemother-apy is effective in achieving satisfactory DFS and OS inchildren with NPC. The use of combined modality treat-ment using chemotherapy and RT has improved DFS andOS in children with NPC. Despite this improvement, about30–40% of children with early-stage disease and 50–60%of children with advanced-stage disease fail to achievelong-term cure. To improve the current results, efforts areneeded to evaluate and develop newer chemotherapeuticregimens to be used in an optimal dose and sequence alongwith RT. Improvement in outcome in the future may beexpected by including concurrent chemotherapy, intensity-modulated RT, biologic response modifiers, and/or immunemodulation. In view of the rarity of the tumor, large collab-orative studies are needed to answer these questions in amore systematic and reproducible manner.

194 I. J. Radiation Oncology ● Biology ● Physics Volume 58, Number 1, 2004

REFERENCES

1. Deutsch M, Mercado R, Persons JA. Cancer of the nasophar-ynx in children. Cancer 1978;41:1128–1133.

2. Ingersoll L, Woo SY, Donaldson S, et al. Nasopharyngealcarcinoma in young: A combined M.D. Anderson and Stan-ford experience. Int J Radiat Oncol Biol Phys 1990;19:881–887.

3. Pao WJ, Hustu HO, Douglass EC, et al. Pediatric nasopha-ryngeal carcinoma: Long-term follow-up of 29 patients. Int JRadiat Oncol Biol Phys 1989;7:299–305.

4. Berry MP, Smith CR, Brown TC, et al. Nasopharyngealcarcinoma in young. Int J Radiat Oncol Biol Phys 1980;6:415–421.

5. Commoun M, Vogt Hoerner G, Mourali N. Tumors of thenasopharynx in Tunisia: An anatomic and clinical study basedon 143 cases. Cancer 1974;33:184–192.

6. Levine PH, Connely RR, Easton JM. Demographic patterns ofnasopharyngeal carcinoma in the United States. Int J Cancer1980;26:741–748.

7. Sham JST, Poon YF, Wei WI, et al. Nasopharyngeal carci-noma: The significance of neck node involvement in relationto the pattern of distant failure. Br J Radiol 1990;63:108–113.

8. Fernandez CH, Cangir A, Samaan NA, et al. Nasopharyngealcarcinoma in children. Cancer 1976;37:2787–2791.

9. Anderson-Anvert M, Forsby N, Klein G. Relationship be-tween Epstein-Barr virus and undifferentiated carcinoma: Cor-related nucleic acid hybridization and histopathological exam-ination. Int J Cancer 1977;20:486–494.

10. Henle W, Henle G. Evidence for an etiological relation ofEpstein-Barr virus to human malignancies. Laryngoscope1977;87:467–473.

11. Bachouchi M, Cvitkovic E, Azli N, et al. High completeresponse in advanced nasopharyngeal carcinoma with bleo-mycin, epirubicin, and cisplatin before radiotherapy. J NatlCancer Inst 1990;82:616–620.

12. Al-Saraf M, Pajak TF, Cooper JS, et al. Chemoradiotherapy inpatients with locally advanced nasopharyngeal carcinoma: ARadiation Therapy Oncology Group study. J Clin Oncol 1990;8:1342–1351.

13. Marcial VA, Pajak TF, Mohiuddin M, et al. Concomitantcisplatinum chemotherapy and radiotherapy in advanced mu-cosal squamous carcinoma of the head and neck. Cancer1990;66:1861–1868.

14. Kim TH, McLaren J, Alvarado CS, et al. Adjuvant chemo-therapy for advanced nasopharyngeal carcinoma in childhood.Cancer 1989;63:1922–1926.

15. Uzel O, Yoruk SO, Sahinler I, et al. Nasopharyngeal carci-noma in childhood: Long-term results of 32 patients. Ra-diother Oncol 2001;58:137–141.

16. Zubizarreta PA, D’Antonio G, Raslawski E, et al. Nasopha-ryngeal carcinoma in childhood and adolescence: A single-institution experience with combined therapy. Cancer 2000;1:89 :690–695.

17. Mertens R, Granzen B, Lassay L, et al. Nasopharyngeal car-cinoma in childhood and adolescence: Concept and prelimi-nary results of the cooperative GPOH study NPC-91. Cancer1997;80:951–959.

18. Ghim TT, Briones M, Mason P. Effective adjuvant chemo-therapy for advanced nasopharyngeal carcinoma in children:A final update for a long-term prospective study in a singleinstitution. J Pediatr Hematol Oncol 1998;20:131–135.

19. Strojan P, Benedik MD, Kragelj B, et al. Combined radiationand chemotherapy for advanced undifferentiated nasopharyn-geal carcinoma in children. Med Pediatr Oncol 1997;28:366–369.

20. Fleming ID, Cooper J, Henson DE, et al. AJCC cancer stagingmanual. Philadelphia: Lippincott-Raven, 1997.

21. Miller AB, Hoogstraten B, Staquet M, et al. Reporting resultsof cancer treatment. Cancer 1981;47:207–214.

22. Kaplan EL, Meier P. Nonparametric estimations from incom-plete observations. J Am Stat Assoc 1958;53:457–481.

23. Peto R, Pike MC, Armitage P, et al. Design and analysis ofrandomised clinical trials requiring prolonged observation ofeach patient. Br J Cancer 1977;35:1–39.

24. Dinshaw KA, Rao DN, Ganesh B. Annual Report 1998. Hos-pital Cancer Registry, Tata Memorial Hospital, Mumbai,2002.

25. Plowman PN. Rare tumors. In: Plowman PN, Pinkerton CR,editors. Paediatric oncology: Clinical practice and controver-sies. Cambridge, UK: Chapman and Hall University Press;1992. p. 446–459.

26. Sham JST, Poon YF, Wei WI, et al. Nasopharyngeal carci-noma: The significance of neck node involvement in relationto the pattern of distant failure. Br J Radiol 1990;63:108–113.

27. Tannock I, Payne D, Cummings B, et al. Sequential chemo-therapy and radiation for nasopharyngeal cancer: Absence oflong-term benefit despite a high rate of tumor response tochemotherapy. J Clin Oncol 1987;5:629–634.

28. Ayan I, Altun M. Nasopharyngeal carcinoma of children ofIstanbul, Turkey. Int J Radiat Oncol Biol Phys 1996;35:485–492.

29. Sahraoui S, Acharki A, Benider A, et al. Nasopharyngealcarcinoma in children under 15 years of age: A retrospectivereview of 65 patients. Ann Oncol 1999;10:1499–1502.

30. Martin WMO, Shah KJ. Carcinoma of the nasopharynx inyoung patients. Int J Radiat Oncol Biol Phys 1994;28:991–999.

31. Lombardi F, Gasparini M, Gianni C. Nasopharyngeal carci-noma in childhood. Med Pediatr Oncol 1982;10:243–250.

32. Labo-Sanahuja F, Garcia I, Xarranza A, et al. Treatmentoutcome of undifferentiated carcinoma of the nasopharynx inchildhood: A 13 years experience. Med Pediatr Oncol 1986;14:6–11.

33. Jenkins DT, Anderson JR, Jereb B. Nasopharyngeal carcino-ma—A retrospective review of patients less than thirty yearsof age: A report from Children’s Cancer Study Group. Cancer1981;47:360–366.

34. Healy GB. Malignant tumors of the head and neck in children:Diagnosis and treatment. Otolaryngol Clin North Am 1980;13:483–488.

35. Baker SR, McCaltchey KD. Carcinoma of nasopharynx inchildhood. Head Neck Surg 1981;89:555–559.

36. Al-Saraf M, Le Blanc M, Giri PG. Chemoradiotherapy versusradiotherapy in patients with advanced nasopharyngeal can-cer: Phase III randomized Intergroup Study 0099. J ClinOncol 1998;16:1310–1317.

37. International Nasopharynx Cancer Study Group. VUMCA Itrial: Preliminary results of randomized trial comparing neo-adjuvant chemotherapy (cisplatin, epirubicin, bleomycin) plusradiotherapy versus radiotherapy alone in stage IV (�N2, M0)undifferentiated nasopharyngeal carcinoma: A positive effecton progression-free survival. Int J Radiat Oncol Biol Phys1996;35:463–469.

38. Kalifa C, Schwaab G, Lemerle J. Les carcinomas indifferen-cies du nasopharynx; problems pediatriques. In: Masson, ed-itor. Actualites carcinologiques Institut Gustave-Roussy. No.1. Paris; 1983. p. 100–105.

39. Azli N, Armand JP, Rahal M. Alternative chemo-radiotherapywith cisplatin and 5-fluorouracil plus bleomycin by continu-ous infusion for locally advanced undifferentiated carcinomanasopharyngeal type. Eur J Cancer 1992;28A:1792–1797.

40. Vita HC, Mendiondo OA, Shaw Dl. Nasopharyngeal carci-noma in second decade of life. Radiology 1983;148:253–256.

195NPC in childhood ● S. LASKAR et al.