IMRT combined with concurrent chemotherapy plus adjuvant chemotherapy versus IMRT combined with concurrent chemotherapy alone in patients with nasopharyngeal carcinoma

Purpose To evaluate the efficacy of IMRT combined with concurrent chemotherapy followed by adjuvant chemotherapy compared with IMRT combined with concurrent chemotherapy alone in patients with nasopharyngeal carcinoma. Methods From January 2007 to December 2014, we collected 797 staged II-IVb [UICC = Union for International Cancer Control criteria (7th edition)] NPC patients for analysis. After 1:1 matching,we selected 261 cases as the CCRT group, another 261 patients as the CCRT+AC group. Using Kaplan-Meier to calculate the overall survival (OS), locoregional failure-free survival(LFFS), distant metastasis failure-free survival(DMFS). The log-rank test and Cox-proportional hazards model to evaluate the prognostic factors. Results After matching, there were 261 patients in each group. In CCRT+AC group, The 1-,2- and 3- year os rates were a little higher than in CCRT group(99.6% vs 97.9%,97.4% vs 96.2%,93.8% vs 86.9%, P = 0.150). There were no significant difference in 1-,2-,3- year OS, LFFS, DMFS between the two groups. In subgroup analysis, a little higher OS rate in CCRT+AC group for staged III, IV and T4(III:100% vs 100%, 97.6% vs 95.8%, 94.0% vs 84.0%; IV: 99.1% vs 95.4%, 96.3% vs 95.4%, 90.5% vs 79.4%, P = 0.047;T4:99.1% vs 95.2%, 97.1% vs 95.2%, 90.9% vs 78.2%, P = 0.055). No significant difference were observed in OS, LFFS,DMFS between the groups. Conclusion IMRT combined with concurrent chemotherapy followed by adjuvant chemotherapy might improved 1-,2-,3- year of OS. Whether or not add adjuvant chemotherapy it had similar LFFS rate and DMFS rate in patients with nasopharyngeal carcinoma. Locally advanced NPC patients (III, IV and T4)might benefit from the adjuvant chemotherapy.


INTRODUCTION
Nasopharyngeal carcinoma (NPC)is popular in China, especially in Guangdong, Guangxi, Hunan, Hainan et al. Radiation therapy is the major treatment for NPC. At present, the standard treatment for NPC is concurrent chemo-radiotherapy. Since 0099 trail [1] reported that conventional chemotherapy combined with concurrent chemotherapy plus adjuvant chemotherapy can improved the overall survival in NPC patients in 1998. This study made sure the efficacy in treated with NPC. Joseph Wee [2] also verified the results as similar as 0099 trail. At the same year, they found that chemotherapy can improve the control rate of distant metastasis . Based on those research, the treatment of conventional radiation therapy combined with concurrent chemotherapy followed by

Clinical Research Paper
Oncotarget 39684 www.impactjournals.com/oncotarget adjuvant chemotherapy became a standard method to treat locoregionally advanced nasopharyngeal carcinoma at North America. But recently, a number of Meta analysis [3,4] showed that, concurrent chemo-radiotherapy plus adjuvant chemotherapy could not improved the survival rate and most of researches were based on conventional radiotherapy. Now coming into the IMRT era, the problem whether or not adjuvant chemotherapy should be given after finished the concurrent chemo-radiotherapy stills controversial. So we collected 522 staged II-IVb NPC patients who received IMRT combined with concurrent chemotherapy followed by adjuvant chemotherapy or only received IMRT combined with concurrent chemotherapy, to investigate the value of adjuvant chemotherapy in NPC treatment.

Patients
We retrospectively collected a total of 797 stagedII-IVb[UICC = (Union for International Cancer Control criteria (7th edition)] NPC patients at the department of radiation oncology of tumor hospital affiliated with Guangxi Medical University, Nanning PR China, from January 2007 to December 2014. Included 261 cases treated with CCRT and 536 patients treated with CCRT+AC. The inclusion criteria:(1) All the patients were initial treatment who never received any chemotherapy or Those who had radiotherapy, chemotherapy or surgery to the head and neck,or those patients who had double cancer, serious heart and lung diseases ,distant metastasis before treatment were exclusion. Before treated , all the patients had a medical history, physical examination, assessment of performance status, complete blood cell count, full bio-chemical profile, EB-virus DNA copies, electrocardiogram, chest computerized tomography (CT), abdominal ultrasound, magnetic resonance imaging (MRI) for nasopharyngeal and neck , nasopharyngeal fiber optic endoscopy and bone scan. In CCRT+AC group ,all the participants were received cisplatin plus 5-fluorouracil (PF) as the pattern of adjuvant chemotherapy .Because the KPS of baseline between the two groups were statistically significant difference . So we selected totally 261 patients who received CCRT , we matched 1:1 with the CCRT group on age, sex, overall stage, T classification, N classification, KPS, as the CCRT+AC group. After matched ,the baseline data of the two groups balance. (Table 1)

Radiotherapy
All enrolled patients were received intensitymodulated radiotherapy(IMRT) technique what was administered by a 6 MV-X ray liner accelerator. Patients were immobilized with a head--neck-shoulder thermoplastic mask in the supine position. Then a CT simulation was carried out to scan the head and neck from the bottom-up. The scan range from the vertex to 2 cm below the clavicle. The slice extending from 2cm over the anterior clinoid process to the level of cartilagines thyreoidea. The slice thickness was 2.5mm, apart from 2.5mm,the remaining were 5mm. We use Precise Plan 2.11(Elekta,Crawley,UK)he Philips pinnacle v8.0(Philips Medical Systems,Milpitas,CA) to delineat the target volumes. The target volumes were delineated according to the International Commission on Radiation Units and Measurements Reports (ICRU)50 and 62.Gross tumor volume in the nasopharynx (GTVnx) stands for what we saw the gross tumor from the clinical examination and imaging examination . Gross tumour volume of involved lymph nodes(GTVnd) was the positive lymph nodes. Clinical tumor volumes one(CTV1) was the high risk clinical tumor volumes , including GTV plus 5-10mm margin and metastasis lymph nodes. Clinical tumor volumes two (CTV2)was the low risk clinical tumor volumes, including the lymphatic regions and CTV1 plus 5-10mm margins. Planning target volumes(PTV) stand for a 3mm margins were added to the GTV and the CTV. The total radiation prescribed dose was 68-74Gy/30-33 fractions to the PGTVnx,60-71Gy/30-33 fractions to the PGTVnd,60-66Gy/28-30 fractions to the PCTV1, 50-60Gy/28-30 fractions to the PCTV2 at 5 fractions per week during a period of 6-7 weeks.

Chemotherapy
Concurrent chemotherapy :1 to 3 cycles of Cisplatin alone 100mg/m2 every 3 weeks; Adjuvant chemotherapy:1 to 3 cycles of Cisplatin 80mg/m2 on day 1 to day 3 and 5-fluorouracil(5-Fu) 750 mg/m2/d on day 1 to day 5 (or continuous intravenous infusion for 120 hours) every 3 weeks (Table 2); But the p value of chemotherapy regimen between two groups had significantly difference ,so we merged patients who receiving 2 cycles with 3 cycles in order to reduce the errors. (Table 3)

Follow-up
After all treatment completion, all patients were subsequently followed up every 3 months for the first two years . Every 6 months through the following 3 to 5 years, and then per annum. The follow-up content including body check, complete blood cell count, liver and kidney founctions, EB virus DNA copies, chest computerized tomography (CT), abdominal ultrasound, magnetic resonance imaging (MRI) for nasopharyngeal and neck , nasopharyngeal fiber optic endoscopy and bone scan. The primary endpoint was OS,LFFS,DMFS. The whole group median follow-up period was 25.65 months( range, 2.13-98.70 months).

Statistical analysis
All data were analyzed by the statistical package for Social Sciences, version 19.0 (SPSS,Chicago, IL.USA). We measured the survival time from the first day the patient diagnosed NPC to the day of the event. Theχ 2 test was used to compare the based-line data(e.g: sex, KPS, clinical stage, T classification ,N classification). For those small sample data we used the fisher's exact test. Using independent-samples T test to compare the age. The Kaplan-Meier method was used to calculate the OS,LFFS and DMFS rates. The log-rank test was used to compare the statistical differences in endpoints in both groups. Using Cox proportional hazards model to calculate the Multivariate analysis. All statistical tests were two-sided, and P<0.05 was considered statistically significant .

DMFS
There were 42 patients had distant metastasis, including 14 patients in CCRT group and 28 patients in CCRT+AC group. Among them, there were 12 cases had distant metastasis in lung,9 cases in bone, 6 cases in brain, bone and lung,5 cases both in lung and bone,5 cases in liver,1 case both in lung and liver,1 case in bone and brain,2 cases in lung, bone and liver (      Table 5).

Multivariate analyses
Multivariate Analyses by Cox proportional hazards model revealed that age, KPS scores, overall stage, treatment method and chemotherapy regimen were not the prognostic factors for OS. Only course of the concurrent chemotherapy was the prognostic factors for DMSF( Table  4).

DISCUSSION
In the age of two-dimensional radiotherapy, the five-year overall survival rate had a good efficiency in NPC [5], but it brought heavier sequelae and more chances to get radio-encephalopathy, patients had low quality of life. Since intensity modulated radiotherapy technique arose, because its targeted volume is closer to the tumor's shape, and it has lower injury to proximal critical normal tissues. So it gradually replace the twodimensional radiotherapy, becoming a principal treatment method for NPC patients. Zhang et al [6] made a study in the application of two-dimensional radiotherapy and IMRT in NPC patients, the results showed that using IMRT can obviously relieved acute or chronic radiation damage, and it had lower chance to have radio-sequelae than using two-dimensional radiotherapy, but IMRT could not improve distant metastasis-free rate, failure-free survival rate and overall survival rate. Later, Lai et al. [7] make a comparison between IMRT and two-dimensional radiotherapy in NPC. They found that IMRT significantly improved the treatment efficacy, patients achieved a good regional control rate, especially for those patients who had early T stage. In 2004, a multi-institutional survey of the effectiveness of chemotherapy plus radiotherapy for NPC by Mitsuhiko Kawashima [8] showed that, concurrent chemotherapy combine with radiotherapy could significantly improved the overall survival rate than only received radiotherapy in advanced NPC patients. A great deal of studies [9,10,20] had approved concurrent chemoradiotherapy had advantages in improving overall survival rate and progression free survival rate, what indicated that locally advanced NPC patients received concurrent chemoradiotherapy would more beneficial. So in 2010, National Comprehensive Cancer Network (NCCN) recommended concurrent chemoradiotherapy was the standard treatment method for NPC. In 2014, Tingting Xu et al [11] make a comparison between concurrent chemoradiotherapy and neoadjuvant chemotherapy for locally advanced NPC patients. The results pointed out both regimens showed similar efficacy. Only concurrent chemotherapy could ameliorate distant metastasis free survival rate for those stage T3-4N0-1 NPC patients. In a trail by su [12], 865 NPC patients received IMRT alone or plus chemotherapy, they classified the candidates into four groups, From the study they made a conclusion that the early disease group had the lowest treatment failure rate. For those NPC patients with early stage, IMRT alone could produce satisfactory results. While for those patients who had locally advanced, chemoradiotherapy was better. Early in 1988, Rossi A et al. [13] had a comparison between radiotherapy plus adjuvant chemotherapy and radiotherapy alone. They found there were no significantly difference in two-year progression-free survival rate and the four-year overall survival rate (55.8% vs. 57.7%, 67.3% vs. 58.5%). In 2010, Kyong Hwa Park et al. [14] reported that the overall response rate of CCRT was 95%, after used AC, the overall response rate was 100%. So they make a conclusion that using cisplatin and 5-FU in combination with radiotherapy followed by three cycles of BEC chemotherapy was effective in locally advanced NPC patients. In 2011,a study by Anne W.M. Lee et al. [15] showed that, concurrent chemoradiotherapy followed by adjuvant chemotherapy can significantly reduced treatment failure and cancer-specific deaths. Recently, a meta analysis by Marie Yan showed that concurrent chemoradiotherapy followed by adjuvant chemotherapy can't enhance survival rates. An other meta-analysis at the same year by Blanchard P et al. [4] found that, concurrent chemoradiotherapy were better than chemotherapy alone in improving survival rates. In our study, no significantly difference in 1-,2-,3-year OS, LFFS and DMFS between CCRT alone and CCRT+AC in NPC patients(χ 2 = 2.072, P = 0.150; χ 2 = 0.363, P = 0.547; χ 2 = 0.875, P = 0.350),what are similar to the results by Dora L.W. et al. [16] in 2004. Nevertheless, after stratification by disease stage, we found a statistically significant difference in OS(P = 0.047). But after we pairwise for each stratum , there were no statistical significance for stage III and IV in two-year os rates (χ 2 = 1.811, P = 0.178; χ 2 = 2.168, P = 0.141). We thought it was due to confounding factors. Although other results from the Stratified analysis didn't make a statistically significant difference , we saw a trend that the CCRT+AC group has higher OS than CCRT group for locally advanced NPC(III,IV and T4)patients from the results .Adding adjuvant chemotherapy for locally advanced NPC patients may benefit. Furthermore ,addition of AC couldn't improved the 1-,2-,3-year LFFS and DMFS for N2-3. On the contrary, it may reduced the 1-,2-,3-year LFFS rate or DMFS rate .This results compared with a study reported by liang [19] at 2014 has some inconsistencies .He reported that concurrent chemoradiotherapy plus adjuvant chemotherapy didn't significantly improve the 2-year OS,LFFS,DMFS, but he found a borderline significant difference in os by CCRT+AC treatment in patients with N2-3 disease(P = 0.052).Analyzed its cause, the main possibility is we enlarged the number of participants. Why concurrent chemoradiotherapy followed by adjuvant chemotherapy may reduced the efficacy for N2-3 patients? One possibility is that for those advanced www.impactjournals.com/oncotarget nodal NPC patients(N2-3), concurrent chemoradiotherapy also have a better efficacy, after chemoradiotherapy, local organizations fibrosis so that with poor blood supply, chemotherapy drugs hardly get into the tumor site, so the plasma concentration beyond the reach of effective concentration. Another possibility explanation is that stage N2-3 patients have poor immunity so they can't tolerant the chemotherapy. In a phase 3 multicentre randomised controlled trial reported by chen et al. [17], concurrent chemoradiotherapy followed by cisplatin and fluorouracil chemotherapy did not improve failurefree survival rate in locoregionally advanced NPC patients. The same conclusion by Shiping Yang [18], concurrent chemoradiotherapy plus adjuvant cisplatin or nedaplatin and 5-fluorouracil chemotherapy didn't significantly improve 3 years OS, LRFS,FFS,DMFS. In 2011, a trail reported by Anne W.M. Lee [21], concurrent chemoradiotherapy was important to locoregional control and survival. Moreover, they though 2 concurrent cycles of cisplatin were enough. Adjuvant chemotherapy using fluorouracil-containing combination can improved distant control. At present, it is uncertain whether concurrent chemoradiotherapy followed by chemotherapy have survival benefit.
Although our trail is a big data retrospective analysis, it has several limitations. The reliability needs more prospective studies to verify the conclusion.

CONCLUSIONS
From this study, IMRT combined with concurrent chemotherapy followed by adjuvant chemotherapy in NPC might improve the 1-,2-,and 3-year OS, but no significant difference were observed between the two groups. There was no statistically significant difference in LFFS and DMFS. For those locally advanced NPC patients(III,IV and T4) might benefit from adjuvant chemotherapy .

COMPLIANCE
WITH ETHICAL STANDARDS