Nasopharyngeal carcinoma (NPC) is rare in pediatric age group; however it is the most common malignancy arising in the nasopharynx in this age group¹. NPC displays a distinct racial and geographic distribution, a reflection of its multifactorial etiology. It is rare in the United States and Western Europe ^{2, 3}. In contrast, it is endemic in southern China. Intermediate risk regions include south east Asia, north Africa and the middle east. The incidence of nasopharyngeal carcinoma worldwide is two- to three folds higher in males compared with females ³.

Pediatric NPC is distinguishable from the adult form for its strong association with Epstein-Barr virus (EBV) infection, undifferentiated histology, the high incidence of loco-regionally advanced disease and the predominance of WHO grade III undifferentiated histology ^{2, 4, 5}.

It is clinically staged according to the International Union Against Cancer (UICC) and the American Joint Committee on Cancer (AJCC) according to Tumor Node Metastasis (TNM) system which is the most important documented prognostic factor⁵.

Patient with early disease is recommended to be treated with radiation-therapy only, while patient with loco-regionally advanced or metastatic disease may receive induction chemotherapy followed by chemo-radiotherapy². Undifferentiated NPC is very sensitive to radiotherapy, which is the mainstay of treatment. With radiotherapy alone, 5-year survival has been reported as 20–60% in most pediatric series ^{6, 7, 8}. Induction chemotherapy has shown a higher response rates in patients with undifferentiated NPC^{9, 10, 11, 12, 13, 14}.

The aim of the present study is to estimate the overall survival (OS) and the event free survival (EFS) of NPC patients treated at Children Cancer Hospital – Egypt (CCHE) during 5.5 years period. Furthermore, common toxicities, the effect of induction chemotherapy regimen followed by concurrent chemo-radiotherapy on survival will be investigated.

This is a retrospective study including all newly diagnosed pediatric NPC at Children Cancer Hospital Egypt ^CCHE during the time period from July 2007 to December 2012. The patients' profiles were systematically reviewed for characteristics including pathology, laboratory, radiological work up and documented toxicities. All imaging studies (e.g., CT and MRI scans) were reviewed by a senior head and neck radiologist for proper staging and assessment of tumor response.

Data was analyzed using IBM SPSS advanced statistical version 20 (SPSS Inc., Chicago, IT). Numerical data were expressed as mean and standard deviation or median and range as appropriate. Qualitative data were expressed as frequency and percentage. Survival analysis was done using Kaplan-Meier method. The comparison between two survival curves was performed using log-rank test. A p-value <0.05 was considered significant. The follow-up time for each patient was the time from enrollment to the closing date for analyses or the date of last information. For overall survival, all deaths were counted regardless of the cause, and the survival times for living patients were censored at the closing date. The first progression at any site or death without progression was counted as an event for DFS. One patient excluded from EFS as he was lost for follow up before post induction evaluation.

Nasopharyngeal carcinoma (NPC) remains one of the greatest oncological challenges because of its highly aggressive natural behavior and the anatomical proximity to critical structures¹. In the young population, it is a rare malignancy, although it is one of the most common epithelial tumors in children. The incidence varies extensively according to racial and geographical factors².

In the present study, the age of patients ranged from 7.8 to 17 years with a median of 12 years. This is relatively in agreement with other studies ^{15, 16, 17, 18, 19, 20, 21}. Our demographic data showed a male predominance, similar to most of the published pediatric NPC series ^{18, 19, 21}.

The majority (84.6 %) of the patients had an advanced ^{stage 3 or 4} disease . This is in agreement with an Indian study reported by Laskar et al. 2008, where stage IV represented 61% of their patients, stage III 27.7% and stage IIb 11% ²². In three other studies, stage IV presentation ranged from 50-56% ^{23, 24, 25}. Late presentation may be explained by the fact that NPC is highly malignant with early lymphatic spread and predilection for distant metastasis. Furthermore, early detection is difficult due to its silent deep- seated location²⁶.

Nineteen of our patients (73%) achieved good response to induction therapy (CR-PR), 5 (19.3%) had SD, while 2 (7.7%) had PD during therapy. The same findings were seen in most of studies ^{20, 21, 22, 23, 24, 26, 27}.

All our patients were histologically WHO type 3, which is in accordance with other studies ^{20, 21, 23}. On the other hand, Daoud et al, reported 56.3% of patients as undifferentiated carcinoma (WHO type 3), and the remainder had a non-keratinising carcinoma (WHO type 2)²⁶.

At the end of our study, 69.2% of the patients were alive in CR, 4 (15.4%) died, 3.8% had PD, while three patients (11.2%) was lost for follow up. Similar results were reported by Venkitaraman et al., and Shen et al.^{21, 22, 23}. Achieving CR at end of treatment was highest in the German Pediatric study (NPC-91-GPOH). With a response rate of 91% where patients were treated by neoadjuvant chemotherapy with (cisplatin, 5 FU, and methotrexate), followed by radiation-therapy and adjuvant interferon β for 6 months²⁷. Similarly, Galindo et al reported good responders to neoadjuvant chemotherapy were 93.75%, and 6.3% of patients had PD²⁴.These results were better than that reported (33.3%) in the Turkish study conducted on 84 Patients. ²⁵.

In a study conducted by Laskar et al., good responders were 47.2%, while 52.8% of cases had SD or PD ²². Another study showed 82.1% of patients had good response, 10.9% had PD and 6% had SD²⁸. Mertens et al reported in their study good responders to neo-adjuvant chemotherapy as 94.9% ²⁹. Similarly, in a multicenter study performed by Ozyar et al., 2006, 96.3% of patients either responded completely or partially to neoadjuvant chemotherapy²⁷.

The 3 year estimated OS and EFS rates were 83.3% and 80.7% respectively, that was similar to what has been reported by other groups ^{28, 29, 30}

In the present study, T stage did not show significant prognostic influence. Similar results were reported by other groups^{21, 25}. However, other reported that T- stage carried a statistically significant influence on event free survival only, with no effect on overall survival^{20, 22}. Zaghloul et al. reported, that the survival worsened with the advancement in T stage ³¹.

The OS for N0-N1 vs. N2-N3 was 88.8% vs.82.3% respectively with no statistical difference. Same results were reported by Shen et al and Zaghloul et al. ^{23, 31}. On the contrary, other studies, reported a better EFS for N0-N1 ^{20, 25, 27} .

Twenty-three patients (88.4%) in the present series were stratified as M0, while 3 patients (11.6%) were M1. Their OS survival was 91.3% versus 33.3%. This was the only prognostic factor identified in our study.

The role of chemotherapy in addition to radiotherapy is becoming the standard of care, and has been proved by many studies ^{32, 33, 34}. Concurrent chemo-radiotherapy²⁷, and adjuvant chemotherapy, both improve OS when compared with RT alone ^{20, 22, 24, 32}.Neoadjuvant chemotherapy failed to improve long term survival as reported by few Chinese studies^{23, 35, 36}. In our study, there was no OS statistical difference between patients according to their response to neoadjuvant chemotherapy. On the other hand, some studies showed that neoadjuvant chemotherapy has better OS, and is better tolerated than adjuvant chemotherapy. Furthermore, early use of a potent combination of cytotoxic drugs at full dose would theoretically be more effective for eradicating micrometastases, shrink the primary tumor, and give a wider margin for irradiation sparing critical neural structures accompanied with less toxicity ^{20, 29, 30, 31} .

On the other hand, controversy exists about the best chemotherapeutic agent to be used. Huang et al. in 2009, showed that concurrent chemo-radiotherapy with carboplatin didn’t improve the survival when compared with induction chemotherapy with carboplatin and 5- fleurouracil ³⁵. No solid conclusion could be drawn pointing to the need for further assessment in larger trials ³⁶. The optimal treatment strategy for NPC patients has not been established through randomized clinical trials yet ^{25, 26}.

The development of RT technique to 3-dimensional (3D) conformal (CRT), and then Intensity-Modulated Radiotherapy (IMRT) not only improved tumor coverage and ensured precision of delineation, but also had better sparing of normal structures and lowered toxicity ²⁶. In our study, all our patients were treated with IMRT with minimal, generally acceptable immediate and late toxicities.

The optimal dose of radiotherapy is still controversial, particularly when combined with chemotherapy. Recently, several studies showed that dose reduction to 60 Gy or less was feasible in multimodality treatment without compromising the outcome in chemo responsive patients ^{21, 22, 25, 27, 30, 37}. On the contrary, several authors reported a better OS with radiation dose > 66 Gy in comparison with lower doses ^{12, 14, 20, 24}.

Neoadjuvant chemotherapy followed by concurrent chemoradiotherapy using IMRT lead to good clinical end results with limited toxicity. Metastatic disease at presentation was identified as the adverse prognostic factor.

Supplementary Figure1

Supplementary Figure2

Supplementary Figure3

Supplementary Figure4