DOI: 10.1093/jnci/djr432 © The Author 2011. Published by Oxford University Press. All rights reserved.

Advance Access publication on November 4, 2011. For Permissions, please e-mail: journals.permissions@oup.com.

ARTICLE

Concurrent Chemoradiotherapy vs Radiotherapy Alone in Stage
II Nasopharyngeal Carcinoma: Phase III Randomized Trial
Qiu-Yan  Chen, Yue-Feng  Wen, Ling  Guo, Huai  Liu, Pei-Yu  Huang, Hao-Yuan  Mo, Ning-Wei  Li, Yan-Qun  Xiang, Dong-Hua  Luo,
Fang  Qiu, Rui  Sun, Man-Quan  Deng, Ming-Yuan  Chen, Yi-Jun  Hua, Xiang  Guo, Ka-Jia  Cao, Ming-Huang  Hong, Chao-Nan  Qian,
Hai-Qiang Mai

Manuscript received April 25, 2011; revised September 15, 2011; accepted September 27, 2011.
Correspondence to: Hai-Qiang Mai, MD, PhD, Department of Nasopharyngeal Carcinoma, Sun Yat-sen University Cancer Center, 651 Dongfeng Rd East,
Guangzhou 510060, People’s Republic of China (e-mail: maihq@mail.sysu.edu.cn).

Downloaded from http://jnci.oxfordjournals.org/ at Univ of Iowa-Law Library on March 14, 2015
Background Concurrent chemoradiotherapy (CCRT) has been shown to improve outcomes for stage III–IV nasopharyngeal
carcinoma (NPC) patients compared with radiotherapy (RT) alone, but the effectiveness of the combined
therapy for stage II NPC patients is unknown.

Methods Patients with Chinese 1992 stage II NPC were randomly assigned to receive either RT alone (n = 114) or CCRT
(n = 116). The CCRT patients were given concurrent cisplatin (30 mg/m2 on day 1) weekly during RT. The
primary endpoint was overall survival (OS). Secondary endpoints were progression-free survival (PFS), distant
metastasis-free survival, and locoregional relapse-free survival. All patients were analyzed by the intent-to-treat
principle. The Cox proportional hazards model was used to calculate hazard ratios (HRs) with 95% confidence
intervals (CIs) and in multivariable analyses to test the independent statistical significance of treatment interven-
tion. Toxic effects and the response to treatment were analyzed using the x2 test. All statistical tests were
two-sided.

Results With a median follow-up of 60 months, adding chemotherapy statistically significantly improved the 5-year
OS rate (94.5% vs 85.8%; HR of death = 0.30, 95% CI = 0.12 to 0.76; P = .007), PFS (87.9% vs 77.8%; HR of
progression = 0.45, 95% CI = 0.23 to 0.88; P = .017), and distant metastasis-free survival (94.8% vs 83.9%; HR of
distant relapse = 0.27, 95% CI = 0.10 to 0.74; P = .007); however, there was no statistically significant difference
in the 5-year locoregional relapse-free survival rate (93.0% vs 91.1%; HR of locoregional relapse = 0.61, 95%
CI = 0.25 to 1.51; P = .29). Multivariable analysis showed that the number of chemotherapy cycles was the only
independent factor that was associated with OS, PFS, and distant control in stage II NPC. The CCRT arm expe-
rienced statistically significantly more acute toxic effects (P = .001), although the rate of late toxic effects did not
increase statistically significantly.

Conclusion Concurrent chemotherapy and radiotherapy is associated with a considerable survival benefit for patients with
stage II NPC.

 J Natl Cancer Inst 2011;103:1761–1770

Nasopharyngeal carcinoma (NPC) is endemic in Southern China study, and some studies advocate the use of CCRT (12,13) or
and Southeast Asia (1,2). Radiotherapy (RT) is the primary treat- neoadjuvant chemotherapy (14). Cheng et al. (12) have shown that
ment modality. Several prospective randomized trials (3–8) and disease-free survival of patients with stage II disease with CCRT is
meta-analyses (9–11) have demonstrated that concurrent chemora- equal to that of patients with stage I disease with RT alone. Their
diotherapy (CCRT) with or without adjuvant chemotherapy is study suggests that CCRT reverses the unfavorable prognosis of
superior to RT alone in the treatment of locoregionally advanced patients with stage II NPC, reducing their risk of failure to that in
NPC. The prognosis of patients with stage I–II NPC is generally patients with stage I disease. Chemotherapy for stage II NPC is
favorable, and this group of patients has largely been excluded not recommended in the guidelines proposed by the Chinese Anti-
from clinical trials of the combined modality treatment. Cancer Association because current evidence has not demonstrated
Although the National Comprehensive Cancer Network has the advantage of CCRT compared with RT alone. It has been
recommended CCRT for stage II NPC, the evidence for its reported (14,15) that chemotherapy is more effective for low-risk
efficacy is weak. The role of adjunctive chemotherapy for stage II patients than for high-risk NPC patients. The primary objective of
patients has not been defined as a primary endpoint by a phase III this study was to determine the overall survival (OS) benefits for

jnci.oxfordjournals.org   JNCI | Articles 1761

However. The technique and dose schedule of RT for the two (DMFS). patients were required to provide written informed consent before CONT E X T S A N D C A V E A T S entering the study. protocol that was consistent with the treatment policy for NPC at the Sun Yat-sen University Cancer Center (21). Stage II disease (T1-2N1M0 or target volume.to 12-MeV electric beam. Supplementary Table 1. electrocardiography. To detect an increase in the 5-year OS rate from domly assigned to radiotherapy alone (RT) or to CCRT. Brachytherapy was not part of the radiation protocol. China) Downloaded from http://jnci. which consisted of the entire tumor with a 2-cm T2N0M0 with parapharyngeal space involvement). adequate usually through two block–shielding lateral-opposing faciocervical renal function (creatinine clearance ≥ 50 mL/min). Limitations Computed tomography of the chest was not performed as part of Pretreatment Evaluation pretreatment evaluation or follow-up. followed by the shrinking field technique (two tory performance status (a score of 0 or 1 using the Eastern lateral-opposed facial fields or smaller faciocervical fields when the Cooperative Oncology Group System). thus reducing the added benefit of chemotherapy. whereas it was 92% for stage I NPC patients (18). patients with Chinese 1992 stage II NPC with the addition of con- current. for the first 40 Gy. adequate portals. between ages margin in all directions. The goal Study design of our study was to achieve equal outcomes for stages I and II NPC In a randomized phase III trial. RT was given five times a week at 2 Gy/d. and acute groups were identical. counts ≥ 4000/µL and platelet counts ≥ 100 000/µL). progression-free survival.5 mg/dL). but there are no data on treatment outcomes Chinese 1992 stage II NPC using RT alone was assumed to be of the combined therapy for stage II NPC patients. distant metastasis-free survival technique. a metastases may not have been detected. All patients underwent conventional RT using a two-dimensional gression-free survival (PFS). a minimum of 202 patients would be Contribution needed. Prior knowledge Treatment of stage III–IV nasopharyngeal carcinoma (NPC) patients Study Design with concurrent chemoradiotherapy (CCRT) has been shown to Based on previous data. some early lung All patients were evaluated by a complete physical examination. a complete blood count with differential count. raphy. Acute toxic effects were higher for Practices Center of the Sun Yat-sen University Cancer Center. and The lower neck was treated through a single anterior field with 1762   Articles | JNCI Vol. after CCRT. pro. The (WHO) types II–III NPC (17). the 5-year OS rate of patients with improve outcomes. Irradiation fields were chosen of the following criteria: biopsy-proven World Health Organization according to the extent of the tumor as evaluated by MRI. patients treated with CCRT. magnetic resonance imaging conventional two-dimensional RT. adequate hematologic function (white blood cell central axis dose. All patients were treated in the supine position. Provincial Center for Drug Clinical Evaluation. The secondary Radiotherapy objectives of this study were to compare the tumor response. received at least 90% of the mid-depth 18 and 70 years. All patients were referred for dental examination before RT. The spinal cord was excluded metastasis. and the posterior cervical Center. Committee of the Sun Yat-sen University Cancer Center. Patients were treated with a uniform RT and late toxic effects of patients between the two treatment arms. 103. 80% to 92% using CCRT at a significance level of 5% and a statistical power of 90% (19). chest radiography.0 (Anhui for the two groups.org/ at Univ of Iowa-Law Library on March 14. patients with stage II NPC were ran. All of the The posterior boundaries of the portals were shifted to the poste- patients were recruited from Sun Yat-sen University Cancer rior margin of the cervical vertebral body. The planning Patients were evaluated using the Chinese 1992 staging system computed tomography (CT) scan was not performed. Issue 23  |  December 7.oxfordjournals. cervix or basal/squamous cell carcinoma of the skin). or prior malignancy (except carcinoma in situ of the from the photon fields after 40 Gy of radiation were administered. weekly cisplatin-based (30 mg/m2) CRT. Wuhu. 80%. The patients (16) (widely adopted in mainland China. were immobilized in a thermoplastic cast. The associated patient registration and randomization Adding chemotherapy to RT statistically significantly improved the procedures were conducted by telephone at the Good Clinical 5-year overall survival. Thus. 2015 (20) was used to generate a random number table. All patients underwent fiber-optic nasopharyngoscopy. and distant metastasis-free survival. arms using blocks of 4 and 6 based on a 1:1 treatment allocation. 2011 . and Epstein–Barr From the Editors virus serology. Megavoltage photons Methods (6 MV) were used to treat the primary tumor and neck lymph Eligibility Criteria nodes. However. a biochemical profile. and satisfac. abdominal sonog- RT may be more effective for locoregional control. a bone scan. late toxic effects were similar New Drug Statistical Treatment software version 8. The exclusion criteria tumor involved the oropharynx or retropharyngeal lymph node) to included previous treatment of NPC. locoregional relapse-free survival (LRRFS). intensity-modulated (MRI) of the head and neck. and a faciocervical radio- available online) and were eligible for this study if they fulfilled all graph was taken using a simulator. This study was approved by the Clinical Research Ethics triangular regions were irradiated by an 8. the presence of a distant limit irradiation of the spinal cord. and all of Implication the included patients were stratified by nodal status classification Combined chemotherapy and radiotherapy treatment also confers (N0 vs N1) and randomly divided into the CCRT or RT-alone survival benefits to stage II NPC patients. to irradiate the nasopharynx and upper neck in one volume hepatic function (serum bilirubin level < 1.

but they were firmed by physical examination. with the Chinese 1992 staging system (Supplementary Table 1. salvage treatments were given to patients after included patients were reclassified as T1.pdf) was used to assess chemotherapy and Results acute radiation toxic effects.org   JNCI | Articles 1763 . An intention-to-treat principle was applied to all of The complete response rate. Because invasion into the nasal cavity is assigned to T2 routinely performed annually or at the time of the clinical sugges. Chicago. were considered in the modeling process were patient age (≤45 phagia because of mucositis pain. chemotherapy. starting on the the last follow-up visit. Strategies to limit the extent of vs >45 years). domly assigned to one of the two study arms. and MRI who refused to participate. once a week during the treatment period. in the AJCC documented relapse or when the disease was persistent. Complete blood counts and used to compare the differences between the two arms (25). All patients received an antiemetic prophylaxis Toxic effects and the response to treatment were analyzed consisting of 5-hydroxytryptamine-3 receptor antagonist plus 20 mg using the x2 test.midline shielding. respectively. and number of chemotherapy cycles (continuous). the involvement of the retropharyngeal lymph node is treatments included reirradiation. calculated by the Cox proportional hazards model (26). Schoenfeld residuals (27). The PFS was defined as the duration from the 60–62 Gy and 50 Gy. nasopharyngoscopy. Four patients in disappearance of all objective evidence of disease. In addition. Complete blood cell counts and biochemical profiles were assessed and a P less than . was 96. cumulative hazard plots estimated for Cisplatin was stopped if creatinine clearance fell to less than the RT and CCRT groups were parallel. The metastatic lymph node–positive and death from any cause or the censoring of the patient at the date of lymph node–negative neck tissues received RT to a total dose of the last follow-up. Tumor response was patients who did not meet the inclusion criteria and four patients evaluated by physical examination. LRRFS. Weight loss during RT is primarily attributable to dys. which was evaluated 3 months after the patients in the analysis. The salvage system. chest radiography. date of each patient’s random assignment to the date of disease progression or the censoring of the patient at the date of the last Chemotherapy follow-up.cancer. first day of RT. 114 patients Tumor response was classified according to the WHO response were randomly assigned to the RT arm. and 116 patients were criteria (23). The two treatment arms were well balanced with respect to least once every 3 months during the first 3 years and then every baseline characteristics (Table 1). RT delays were strongly discouraged.0 (SPSS. Chinese 1992 node stage supportive medications. The blood chemistry were checked before each chemotherapy cycle. classified as N1 and the involvement of the bilateral neck lymph nodes is classified as N2.org/ at Univ of Iowa-Law Library on March 14. Thus. Kaplan–Meier survival curves were used to of dexamethasone. The LRRFS and DMFS were also evaluated and calcu- Patients who were randomly assigned to the CCRT arm were lated from the date of each patient’s random assignment until the scheduled to receive 30 mg/m2 cisplatin in 1 L of normal saline day of the first locoregional or distant relapse or until the date of over 2 hours on a weekly basis during external RT. available online) but to T1 with the AJCC staging system. IL). The remaining 230 patients were ran- of the head and neck at 3 months after the completion of RT. mucositis and to manage its symptoms included basic oral care and Chinese 1992 tumor stage (T1 vs T2). sex. Nasopharyngoscopy. hazard ratios (HRs) with 95% confidence interval (CI) were Dose modification for cisplatin during CCRT was not allowed. DMFS. the CCRT arm did not receive any chemotherapy. assumption of proportional hazards was appropriate. Late radiation toxic effects were Patient Characteristics assessed using the Radiation Therapy Oncology Group and We assessed 236 patients for eligibility from October 2003 to European Organization for Research and Treatment of Cancer September 2007. In addition. Potentially important prognostic factors that patients. and included in the analysis according to the intention-to-treat prin- MRI. Multivariable analyses were performed using the Cox proportional hazards Management of Mucositis model (26) to test the independent statistical significance of treat- Radiation-induced mucositis is a common toxic effect for NPC ment intervention. gov/forms/CTCAEv3. and surgery. The accumulated radiation dose to the primary from the date of each patient’s random assignment to the date of tumor was 68–70 Gy. 2015 at least 1500/µL and the platelet count was at least 100 000/µL. parapharyngeal space involvement (yes or no). prochlorperazine and lorazepam analyze the time-to-event endpoints (24). direct nasopharyngoscopy. seven patients were reclassified as Statistical Analysis N1 and 31 were reclassified as N2 and stage III.0 scale (ctep. Patients were restaged according 6 months thereafter until death. and the incidence of Response toxic effects. and abdominal sonography were system (28). The OS rate was the primary endpoint of this study.5% (110/114) for the RT arm and jnci. MRI of the to the American Joint Committee on Cancer (AJCC) staging head and neck. with the and cisplatin was delayed until the absolute neutrophil count was assumptions of proportional hazards confirmed based on Downloaded from http://jnci. including two late radiation morbidity scoring schema (22).oxfordjournals. 25 of the When possible. The OS was defined as the duration the completion of RT. which was con. verifying that the 50 mL/min. and the log-rank test was antiemetics were added as needed. Secondary endpoints included PFS. All statistical tests were two-sided. The National Cancer Institute Common Toxicity Criteria version 3. Analyses were performed using the statistical software package Patient Assessment and Follow-up SPSS 16. Six patients were excluded. (N0 vs N1).oxfordjournals.05 was considered to be statistically significant. In total. After the completion of treatment. A complete response was defined as the complete randomly assigned to the CCRT arm (Figure 1). patients were evaluated at ciple. tion of tumor relapse.

In the RT patients given CCRT as compared with RT alone (12. two patients had residual neck lymph nodes and two had P < . The residual neck lymph nodes in the RT arm when evaluated incidence of grade 3–4 mucositis was statistically significantly 6 months after the completion of RT. of their RT course. and six patients Patterns of Treatment Failure (5. eight developed liver metastases. as opposed to none in the CCRT arm (Table 3). P = . Patients with stage II nasopharyngeal carcinoma CCRT arm RT alone arm (Chinese 1992 staging system) were randomly Downloaded from http://jnci.4%) received at least six cisplatin cycles.1% (115/116) for the CCRT arm. With respect to gastrointestinal reactions.oxfordjournals. and five developed distant lymph cantly higher than in the RT arm (63.6% vs 32. the incidence of grade organ.4%.org/ at Univ of Iowa-Law Library on March 14. Assessed for eligibility (n = 236 ) Excluded (n =6 ) Not meeting inclusion criteria (n =2 ) Refused to participate (n =4 ) Other reasons (n = 0 ) Patients randomly assigned (n =230 ) Figure 1.6% vs 0%. The reasons for withdrawal of cisplatin 22. the incidence residual nasopharyngeal tumors.001).35).9%) received a seventh dose. He successfully received higher in the CCRT arm than in the RT arm (45. 2011 . prolonged severe CCRT groups.9% vs 0%. 2015 assigned to the RT alone arm or the CCRT Allocated to intervention (n =116 ) Allocated to intervention (n =114 ) arm. Received allocated intervention (n =112 ) Received allocated intervention (n =114 ) RT = radiotherapy.  CONSORT flow diagram. node metastases. One patient in the CCRT arm of grade 3 nausea or vomiting was statistically significantly higher had residual neck lymph nodes.001). There were no statistically significant differences between the Toxic Effects and Compliance two study arms in the cumulative incidence of grade 3 or higher All patients in both treatment arms completed the prescribed dose late radiation morbidity during follow-up (Table 2). acute toxic effects in the CCRT arm was statistically signifi. The main grade Three patients had both locoregional failures and distant failures 3–4 acute toxic effects during CCRT were hematologic and in the RT alone arm. Ninety-one patients (78. respectively.8% vs 40. 103. No grade 5 toxicity (death) Among those patients who developed distant organ metastases. 1764   Articles | JNCI Vol. arm. P = . P = .04). In the CCRT arm.001). two patients had grade 4 mucositis. CCRT = concurrent chemoradiotherapy. of RT. Did not receive allocated intervention (n =4 ) Did not receive allocated intervention (n =0) Lost to follow-up (n =1 ) Lost to follow-up (n =0 ) Discontinued intervention (n =0 ) Discontinued intervention (n =0 ) Analyzed (n =116 ) Analyzed (n =114 ) Excluded from analysis (n = 0 ) Excluded from analysis (n =0 ) 99. neutropenia.2%) received an eighth dose because of the timing and duration After a median follow-up of 60 months (range = 5–87 months). Some patients had metastases in more than one With respect to hematologic toxicity. There was only one patient with with CCRT than with RT alone (8.5%. There were no statistically 3 leukopenia/neutropenia was statistically significantly higher in significant differences between the two arms (P = . Thirty patients received salvage treatment for relapse. neck lymph node resection.2% (13 of 116) of patients in the RT and included the patient’s refusal. gastrointestinal reactions (Table 2). The overall incidence of grade 3–4 nine developed bone metastases.8% (26 of 114) and 11. Issue 23  |  December 7. 30 patients (25. developed tumor progression (Table 3). occurred during treatment. and cisplatin-induced renal toxicity. five developed lung metastases. severe mucositis.

this study is the first randomized AJCC N stage† trial to compare CCRT with RT alone in early-stage NPC.29). one received brachytherapy.4) for stage II NPC patients seems reasonable. The Total 114 116 Age. who CCRT arm) were disease related (Table 4). The outcomes for this group after RT.5) survival rate equal to those with stage I disease given RT alone   No 23 (20.7) and distant control (HR of metastasis = 0.9% vs and stage IIB disease.2) Hemoglobin level.5) metastases in only those patients with T1-2N0-1M0.35). Song et al.org   JNCI | Articles 1765 . isolated distant than in the RT arm (OS: 94. HR of distant relapse = 0. followed by RT.8) results of neoadjuvant chemotherapy and RT with RT alone for   >20 and <40 20 (17. y addition of cisplatin-based chemotherapy to RT resulted in 8.8%.5) 20 (17.01) in stage II NPC (Table 5). Figure 3.2) 97 (83.25 to 1.6) 105 (90. (14) reported the results of a pooled analysis of Maximum lymph node size.5) 19 (16. P = . chemotherapy plus reirradiation. patients who had stage II disease had worse outcomes than patients who had stage I disease (34). In addition.017. P = . however.8) 19 (16. HR of locoregional relapse = 0. Neoadjuvant chemotherapy showed no additional benefit compared with treatment with RT alone.8) 98 (84. and one received RT alone have been moderately satisfactory. 95% CI = 0.007. PFS: 87.4) Parapharyngeal involvement disease who were treated using CCRT exhibited a disease-free   Yes 91 (79.1) 8 (6.4) 11 (9. one received stereotactic 90% with RT alone (30–33).1) 8 (6.74.9) 108 (93. Figure 2.4) AJCC 1997 stage I–II NPC has been investigated by Cheng et al. 95% CI = 0. DMFS: 94.   WHO type 3 110 (96.oxfordjournals. 95% CI = 0.14. P = . respectively. The results demonstrated that neoadjuvant   >130 101 (88.5) 111 (95.6) AJCC 1997 stage I–II disease.007. Sex A multivariable analysis showed that the number of chemo-   Men 84 (73. PFS.7% and 14. one received neck dissection. metastases occurred in 5. Patient characteristics* Figure 2. of patients Characteristic (%) (%) and RT groups (LRRFS: 93. HR of progression = 0.89.04).4)   T2N1 89 (78.61. demonstrated that after RT. evaluated the treatment results of 1070 NPC patients mostly and DMFS were statistically significantly higher in the CCRT arm treated with RT alone between 1990 and 1998. the above studies were all retrospective analyses (12.5) 19 (16. 95% CI = 0.51. In one report by Leung et al.1%. Chua et al. 12 patients received chemotherapy. Specifically. g/L different disease stages. (14).7) therapy cycles was the only independent factor that was associated   Women 30 (26.76.1) 89 (76.93. Thus.99.   T2 93 (81. Notably. the recommenda- AJCC T stage† tion of CCRT by the National Comprehensive Cancer Network   T1 21 (18. Figure 3. A. which supports the important role of concurrent   T2N0 18 (15. The   N0 13 (11.3) 34 (29.80 to 0.3) PFS (HR of progression = 0.007).2) 18 (15. RT alone CCRT 95% CI = 0. patients with stage II   III 12 (10.9) with RT alone.9)   T2 107 (93.9% increases in the 5-year OS. P = .82. A). (33).10 to 0. (29) compared the out- * AJCC = American Joint Committee on Cancer. B.3) Pathology with OS (HR of death = 0.67 to 0.4) 17 (14.1%.7) chemotherapy for stage II NPC patients.88. In contrast to the studies by Chua et al.29.1) Chinese 1992 N stage Discussion Downloaded from http://jnci.23 to 0.6) To the best of our knowledge. RT = radiotherapy. HR of death = 0. a study nasopharyngectomy.5% vs 85. P = .5) 97 (83. 95% CI = 0.7%.9% of patients with stage IIA 95% CI = 0. in the CCRT arm.96. those for 29 patients following neoadjuvant chemotherapy † Defined by the criteria of the seventh edition of the AJCC staging system (28). AJCC stage group† (12). P = .5) 5 (4. respectively.9%. mm two randomized controlled trials that compared the treatment   ≤20 94 (82. PFS. B).1) 80 (68.6) 97 (83. of which 21 (16 in the RT alone arm and six in the following RT (33. 2015   N0 18 (15. and DMFS   Range 28–70 26–65 rates.8) 19 (16. The 5-year OS.45. and 10. nine received numbers of patients therein were small.69 to Chinese 1992 T stage 0.oxfordjournals.   Median 43 42 10.7) 82 (70.   WHO type 2 4 (3. who reported the treatment outcomes of 44 patients with   II 102 (89.5) 96 (82.5) (12).30. reported that patients with T1-2N1 disease had a relatively poor jnci.6) This randomized trial demonstrated substantial survival benefits of Chinese 1992 stage group CCRT for patients with stage II NPC as compared with treatment   T1N1 7 (6.8% vs 83.79. several studies have 77.27.12 to 0.4) 19 (16. of patients No.8%. P = . It is Survival recognized that the risk of distant failure experienced by some Twenty-five deaths (19 in the RT arm and six in the CCRT arm) patients with stage IIB NPC approaches that of stage III disease were reported.5) chemotherapy improved survival and reduced the risk of distant   ≤130 13 (11. There was no statis- tically significant difference in LRRFS rates between the CCRT No.9) possible impact of combined modality treatment for patients with   N2 12 (10.Table 1.org/ at Univ of Iowa-Law Library on March 14.0% vs 91. and the Among them.   T1 7 (6. CCRT = concurrent comes for 31 patients with early-stage NPC given RT alone with chemoradiotherapy.4)   N1 96 (84. five received traditional Chinese The OS rate for early-stage patients is approximately 80%– medicine. WHO = World Health Organization.7)   N1 89 (78.

2% for the T2N1 group.9) 8 (6. (36) demonstrated that the 5-year accumulated distant lower distant failure rate than the RT arm.3) 109 (94.4) 0 72 (62.04   Skin reaction 10 (8. such as combined have a smaller distant tumor bulk that is more easily eradicated by modality treatment. (6) chemotherapy on NPC patients.36.6) 0 .7%. It is possible that con- metastasis rate of T1-T2N1 (Chinese 1992 staging system) NPC current chemotherapy.57   Total (any) 11 (9. or 4) multiple neck fered from those of the other groups.0) 0 11 (9. 2) supraclavicular distant metastasis rate was 21.47 Downloaded from http://jnci.7) .9) Locoregional only 9 (7.9) 0 . eradicating distant micrometastases in early-stage NPC. Primary outcome of treatment* Primary outcome.8) 0 13 (11. (%) Alive 95 (83. (14) studied the effect of induction vival in patients with stage II NPC is encouraging.9) 2 (1.6) 0 1 (0.31   Renal impairment 0 0 1 (0.7) * CCRT = concurrent chemoradiotherapy. Xiao et al.org/ at Univ of Iowa-Law Library on March 14.1) Distant only 14 (12.4) Site RT alone (n = 114) CCRT (n = 116) Death 19 (16. Two (37) revealed that the 5-year OS rate of patients with T2N1 disease studies (14. This study demonstrates similar vs RT: 87.8) 3 (2. The results showed that statisti- reported that CCRT with daily low doses of oral uracil and tegafur cally significant differences in the OS and DMFS rates were only statistically significantly decreased the incidence of distant metas.Table 2. P < .2) 25 (10. (15) divided NPC (Chinese 1992 staging system) was only 73. which dif. They concluded that distant node metastases with one node greater than 4 cm. which differs sta. Issue 23  |  December 7.9) 2 (1. node metastases. outcome and that more aggressive therapy. high-risk patients.9) 1 (0. 2011 . 1766   Articles | JNCI Vol. Maximum acute and late toxic effects* Toxic effects.3) 21 (9. * CCRT = concurrent chemoradiotherapy.0) 204 (88.5) 0 51 (43. with distant metastasis (14. Lin et al. are for the difference in the incidence of grade 3 and 4 adverse events between the two treatment arms. the CCRT arm had a Zong et al. 3) 1992 AJCC stage T4N2.8) 0 . calculated with the x2 test. patients into high-risk and low-risk subgroups according to their tistically significantly from those of the other groups of early-stage grading system.9)   Disease progression 16 (14. in the low-risk group.9) 4 (1. 103. All statistical tests were two-sided.6) 1 (0.5) 1 (0.32   Liver dysfunction 1 (0.7) Lost to follow-up 0 1 (0. High-risk patients met at least one of the fol- NPC patients.7) . and dose intensity used in the present analysis.66   Trismus 1 (0. † P values. RT = radiotherapy.1) 2 (1.1%. RT = radiotherapy. RT = radiotherapy. may be indicated for such patients (34. P = .9) .37). Chua et al.0) 5 (4.6) 0 15 (12. concurrent chemotherapy.2) 0 .9%) accounting for most of the better in patients receiving CCRT than those receiving RT alone treatment failures in the RT arm.9) 0 0 0 .9) 1 (0.53   Weight loss 3 (2. No.7) 6 (5.001 Late toxic effects   Ear (deafness/otitis) 8 (7.3) 5 (4. No. no survival benefit was gained for Our observation that concurrent chemotherapy improves sur.9) 0 .001   Mucositis 37 (32. How can the distinct effect of concurrent chemotherapy on stage II NPC patients be explained? We speculate that early-stage disease may Table 4.6) 0   Other cause 3 (2. Xiao et al. is more effective in resulting in a much lower OS rate in the former group.32 * CCRT = concurrent chemoradiotherapy.001) were statistically significantly results. at least with the chemotherapy regimens patients was much greater than that of T1-T2N0 NPC patients.7) . (%) RT alone CCRT All Table 3.6) . (%) RT alone (n =114) CCRT (n =116) Toxicity grade Toxic effect 3 4 3 4 P† Acute toxic effects   Leukopenia/neutropenia 0 0 15 (12.3% vs 61.2% vs 59. They found that metastasis after curative RT was the main reason for treatment the OS (CCRT vs RT: 83.3)   Treatment related 0 0 0 Locoregional and distant 3 (2. Incidence and site of first progression* Survival status (n = 114) (n = 116) (n = 230) Incidence.32   Nausea/vomiting 0 0 10 (8. Kwong et al.5%. observed in the T1-T2N0-N1 group and favored the combined tasis and increased survival in advanced-stage NPC.001   Thrombocytopenia 0 0 1 (0.15) support this opinion. however.004) and PFS (CCRT failure in the T2N1 group.9) 0 <.60   Total (any) 46 (40. 2015   Skin fibrosis 2 (1.oxfordjournals. No. In this study.9) . (37) also found that the accumulated lowing criteria: 1) nodal size greater than 6 cm.

007. all patients with quacy of the two-dimensional RT technique. However.39). Downloaded from http://jnci. two-sided log-rank test). recurrent. We found that the static NPC compared with non-cisplatin regimens (38.9% optimal combination schedule of cisplatin and RT has not yet been increase in the 5-year DMFS rates. chemotherapy and RT arm. leading to better Chinese 1992 stage II NPC were low-risk patients according to the outcomes.org   JNCI | Articles 1767 . A) Overall survival (HR of death = 0. two-sided log-rank test). 95% CI = 0. concurrent cisplatin however.45.org/ at Univ of Iowa-Law Library on March 14. Kaplan–Meier estimates of patients who were randomly assigned to RT vs CCRT. B) Progression-free survival (HR of progression = 0.88. CCRT = concurrent chemoradiotherapy. daily low-dose. P = . 2015 Figure 2. RT = radiotherapy. Toxic effects are considerable jnci. weekly intermediate-dose. 95% CI = 0. or meta- more effective in eradicating distant tumors. and chemotherapy was higher response rates in previously untreated. substantial improvements in OS. CI = confidence interval.oxfordjournals. Stage II Cisplatin-based chemotherapy has been shown to result in NPC had a smaller distant tumor bulk.23 to 0. P = .017. The rate with concurrent cisplatin chemotherapy might translate into corresponding 5-year DMFS rates were 86% and 74% (P = .oxfordjournals. grading system previously mentioned by Lin et al.005). HR = hazard ratio. The large reduction of the distant metastasis combined arm and 67% in the RT-alone arm (P = . In this study.30.12 to 0.048). which suggests that concurrent established.76. The 5-year OS was 79% in the radiosensitization. The addition of cisplatin-based chemotherapy to RT showed a 10. (15). and 3-week cisplatin chemotherapy has systemic cytotoxic action in addition to high-dose regimens have been used. the improved outcomes did not appear in the advanced chemotherapy might also compensate for the dosimetric inade- NPC group in their study (14).

 Kaplan–Meier estimates of patients who were randomly assigned to RT vs CCRT. RT = radiotherapy. Issue 23  |  December 7. HR = hazard ratio. We believe that the excellent distant control that was ever. P = . with the standard chemotherapy regimen of cisplatin at 100 mg/m2 and neck (40). Chan et al. patient compliance was unsatisfactory because only 44% of established in this series is attributable to the good compliance patients actually completed six cycles of chemotherapy during RT. therapy. P = . two-sided log-rank test). 95% CI = 0.51. B) Locoregional relapse-free survival (HR of locoregional relapse = 0. Downloaded from http://jnci. This study had good compliance for the mediate dose of cisplatin (40 mg/m2) improved the survival rate as CCRT regimen. CCRT = concurrent chemoradiotherapy.007. mising local control.27. two-sided log-rank test). Kim et al. 2011 .25 to 1.4% completed at least six cycles of chemo- compared with RT alone in locoregionally advanced NPC. 103. (4) reported that CCRT using a weekly inter. (41) reported that a weekly intermediate every 3 weeks during RT. A) Distant metastasis-free survival (HR of distant relapse = 0. resulting in decreased interruptions in radia- receive three courses of concurrent 100 mg/m2 cisplatin actually tion treatment and minimal acute toxic effects without compro- did so. 2015 Figure 3. We attribute good compliance with CCRT in this study to 1768   Articles | JNCI Vol. 78. dose of cisplatin (30 mg/m2) is practical and feasible for the CCRT (3) revealed that only 63% of patients who were scheduled to treatment of NPC. CI = confidence interval. how. A multivariable analysis also showed that Weekly cisplatin at a dose of 30 mg/m2 decreases toxic effects the number of chemotherapy cycles was the only independent without compromising tumor control in patients who receive factor that was associated with OS and distant control in stage II CCRT for locally advanced squamous cell carcinoma of the head NPC.74. 95% CI = 0.oxfordjournals.org/ at Univ of Iowa-Law Library on March 14.29. The intergroup study by Al-Sarraf et al.61. with the CCRT regime.10 to 0.

Chin J Cancer. jnci. Tung SY. Ngan RK. eight randomized trials and 1753 patients.14) . et al.81 to 1. Cheng SH.89 (0. Cao SM.88 therapy for nasopharyngeal carcinoma: a factorial study.51 to 3. Randomized trial of radiotherapy versus * Categorical variables were patient age (≤45 vs >45 years).79 (0. of chest x-ray is comparatively low for the detection of lung metas.33 survival. Progression   Sex 0.56 to 3.23(27):6730–6738.47 to 2.21(4):631–637. hypotheses of this study was that the outcome in terms of survival 2000. Int J Radiat Oncol Biol Phys.org/ at Univ of Iowa-Law Library on March 14. Tai BC.75) .org   JNCI | Articles 1769 . of chemotherapy cycles 0. Lau WH. Leemans CR. and mucositis acute toxic therapy alone in locally advanced nasopharyngeal carcinoma: results of a effects than the patients undergoing RT alone. 2015   Sex 0.47 to 3. gastrointestinal. 2002.48) . Concurrent chemotherapy-   T stage 0.66 (0.   N stage 1. LeBlanc M. et al.94 (0. Am J Clin Oncol. Chan AT. 8.34 stage II NPC. et al. First. Phase III study   Sex 1. In summary. Berkhof J. we think that the optimal 10.81) .31 (0.09 (0.81 (0.46 (0. Bourhis J. Cheng SH. The prevalence and prevention of naso-   T stage 1. Wee JT.18(10):2040–2045. Because intensity-modulated RT (IMRT) 15. Ma J. There were a few limitations to this study. Another way to estimate for the treatment of NPC tends to be more effective in locore.55 6. J Clin Oncol. 2004. study on therapeutic gain by concurrent chemotherapy for regionally- advanced nasopharyngeal carcinoma: NPC-9901 Trial by the Hong Kong Nasopharyngeal Cancer Study Group. HR = hazard ratio. J Clin tumors and to ensure NPC patient compliance.24 to 1. Chinese 1992 tumor stage (T1 vs T2). Giri PG.44). Lin AC. Ha TC.43 radiotherapy in patients with advanced nasopharyngeal cancer: phase III   Age 1.78 to 59.71 (0. et al. tional value of chemotherapy to radiotherapy in locally advanced nasopha- for both an optimal chemotherapy effect to eradicate small distant ryngeal carcinoma: a meta-analysis of the published literature.47) .528 patients from six randomized trials. Is nasopharyngeal cancer really a   No.30(2):114–119. Liang WM. Distant failure 3.44 nasopharyngeal carcinoma: positive effect on overall and progression-free   Parapharyngeal space involvement 0. J Clin Oncol.82) . 12.67 to 0.22(22):4604–4612. Concurrent and adjuvant chemo-   N stage 1. Oncol.84) . Sham JS.83) .   No. of induction chemotherapy to radiotherapy in patients with early-stage nasopharyngeal carcinoma: subgroup analysis of two phase III trials. 14. Chemotherapy in locally advanced nasopharyngeal carcinoma: an individual patient data meta-analysis of not part of the pretreatment evaluation and follow-up schedule. Teo PM.58 to 3.99) .90 (0. Qian CN.64 of concurrent chemoradiotherapy versus radiotherapy alone for advanced   Age 1.42 to 4.007 II NPC when IMRT is used. 5% weight loss.77   No.oxfordjournals.33) . Int J Radiat Oncol evaluation and follow-up of NPC patients because the sensitivity Biol Phys. inal ultrasonography.76 to 3. the margin of benefit gained with additional diotherapy adequate? Int J Radiat Oncol Biol Phys.89) . geal space venous plexus and marrow involvement: potential landmarks of From this point of view. this randomized study demonstrated several Death (all causes) distinct and substantial survival benefits of CCRT for patients with   Sex 0. The addi- choice for early-stage NPC is cisplatin.10) .80 to 0. et al.01 randomized trial.67) . Slotman BJ. Tsai SY. dimensional technique. Baujat B. variety. J Clin Oncol. Lee AW. J Clin Oncol. J Clin Oncol. Al-Sarraf M. they were all meta-analysis of 1. Endpoint HR (95% CI) P† In conclusion. Lin JC. Sham JS. Jiang RS.57 (0. 2005. 2002.69 to 2. 2004.26 to 5.22 to 16. Chua DT. Locoregional control 5. CI = confidence interval. Buter J.80   Age 1. Yen KL.82 (0.85 is warranted to explore the role of CCRT in the treatment of stage   N stage 0.16(4):1310–1317. Loong SL. 2006.18 treated with IMRT in China since 2009 because of improving   Parapharyngeal space involvement 3.41 (0. 2005. Hsu CY. Au GK.33 (0.23 to 1. and skeletal scintigraphy is routinely per. Improvement of survival after addition tases.11 medical resources. Concomitant radiotherapy and formed to detect distant metastases in NPC (42).69 to 0. Lin JC.65(5):1300–1306.64(1):47–56. Conventional medical evaluation with chest radiography.35) . 2006. Int all patients in this study underwent conventional RT using a two- J Radiat Oncol Biol Phys.60(1):156–164. Tsai SY.04 “Cantonese cancer”? Chin J Cancer.47) .71 radiotherapy compared with radiotherapy alone in locoregionally advanced   N stage 1. chest CT should be included as part of the dissemination for stage I-III nasopharyngeal carcinoma. Kwong DL.55 nasopharyngeal carcinoma: progression-free survival analysis of a phase III   No.25 (0.93) . Multivariable analysis of prognostic factors in Chinese chemotherapy in stage II NPC may be reduced when IMRT is 1992 stage II nasopharyngeal carcinoma* used. tolerant of this regimen. Second. Kupelnick B. Number of chemotherapy cycles was against cancer stage III and IV nasopharyngeal cancer of the endemic continuous. Chemoradiotherapy versus Downloaded from http://jnci.93) . Langendijk JA. a chest CT was 11.55) .19 to 1.61(2):456–465.78 pharyngeal carcinoma in China. † P values were calculated using the two-sided Wald test in the Cox propor.96) . 1998.44 (0. 2004.23(28): admission for intravenous nutrition when the patients developed 6966–6975. Wang WY. 2010. of chemotherapy cycles 0.44 to 1.20(8):2038–2044. because of a lack of medical resources in China. at a weekly dose of 30 mg/m2.31 to 3. Yen KL.29(5):517–526.09 (0. Further investigation of this prospective setting   T stage 0. Tan EH.79) . One of the chemotherapy for early-stage nasopharyngeal carcinoma.31 to 2. of chemotherapy cycles 0. 2003.83 randomized Intergroup study 0099.22) . Audry H.42 2. 2005. J Clin Oncol. Wee J.82 (0.46 (0.86 (0. Combined chemoradiation versus radiation severe hematologic. Preliminary results of a randomized tional hazard model.79) . Huncharek M. Jan JS. parapharyngeal concurrent chemoradiotherapy followed by adjuvant chemotherapy in space involvement (yes or no).25) .66 (0.73 to 16.oxfordjournals. 13. abdom.   Parapharyngeal space involvement 6.09 to 6.25(3):219–223. Jan JS.   T stage 1. Simons MJ. and patients with American Joint Committee on Cancer/International Union Chinese 1992 node stage (N0 vs N1).55 1. Liang WM. et al.07 to 5. Jiang RS.49 22(13):2643–2653.Table 5. 2011.89) .39 References   Parapharyngeal space involvement 1.91 (0. J Clin Oncol. 7. outcome of advanced nasopharyngeal carcinoma—is concurrent chemora- gional control (43. Qian CN. et al. of chemotherapy cycles 0.08 4. et al.62 (0. Prognostic significance of parapharyn- would improve by reducing the occurrence of distant metastases.24 (0. et al. An increasing number of NPC patients have been   Age 1. Although CCRT patients experienced more 9.

-Q. Funding American Joint Committee on Cancer: AJCC Cancer Staging Manual. Cancer. University Cancer Center. Lu TX. Ko YH. New York. P-YH. Wen. patients with clinically localized undifferentiated nasopharyngeal carci- 23. Huang EY.M. Harris J. Lu TX. 36. Biol Phys. 34. and the Fundamental Research Funds for of external plus intracavitary radiotherapy on stage I-II nasopharyngeal the Central Universities. of Guangdong Province (2007B060401064 to H.27(22): 25. Sample Size Tables for Clinical 39. eds. Prospective study on long-term efficacy Research 5010 Program. the Sci-Tech Project Foundation of Guangzhou City geal carcinoma patients treated with conventional external beam radio- (2007Z1-E4022 to H. article.53:457–481. Treatment outcomes for radiotherapy alone are comparable with neoadjuvant chemo- (2006AA02A404 to H. 2009. 40.-F. A study of the combined treatment strategy Y-FW. Park CI.M. Han F. Sze WK. Treatment outcome after radio.11(1):70–76. and Treatment of Cancer (EORTC). 43. et al. Partial residuals for the proportional hazards regression model. Leung SF. Huang Y. 1997. Choy D. Peto R. carcinoma. University Cancer Center. 1991. 2009. Cancer.80(3):661–668.2(2):137–149. to H.30(5):1037–1042. Cancer. 44. 1770   Articles | JNCI Vol. Hong M. Dugan M. Sung MW. study. Chen. Tang L. We also thank professor Qing Liu 33. Sobin LH. Amdur RJ. Morris CG. 2005.27(7):555–565. NY: Springer. A review of the Princess Margaret Hospital experi- 20. Wu SX. 2011 . FQ. for patients with nasopharyngeal carcinoma based on the analysis of the Y-JH. no potential conflicts of interest to declare. 32. Yeh SA. and H. 1997. Min HQ. Cox JD. Chin J Oncol. of the nasopharynx. Guo X. D-HL. WM. People’s Republic of China (Q-YC. Tung SY. et al. the collection. Sun Yat-sen University Cancer Center. Meier P. XG. Pike MC. A new staging system for nasopharyngeal 37. 42. 2003.98(2):288–291. Sun RY. 1972. 1991. Downloaded from http://jnci. Sun Yat-sen treatment results from 749 cases. Zhao C. Int J Radiat Oncol Biol Phys. Guangzhou. HL. Compton CC. et al. This work was supported by the National Natural Science New York. 35.M. Project 863 29.26(2):204–207. Int J Radiat Oncol Biol Phys. RS. Laryngoscope. Lu TX. Oxford. Chen CY. Edge SB. 2008. Treatment outcomes Notes and late complications of 849 patients with nasopharyngeal carcinoma treated with radiotherapy alone. 2000. 27. Luo W. Retrospective analysis of 934 nasopharyn. Lee N. Garden AS. Tannock I. Cancer Res Treat. Lui CC. 2005. Int J Radiat Oncol Biol Phys. Ye WJ. Int J Radiat Oncol Biol Phys. the noma. Head Neck. Reporting results of noma. Hoogstraten B. Chin J Cancer. Pretherapy quantitative measurement Affiliations of authors: State Key Laboratory of Oncology in South of circulating Epstein-Barr virus DNA is predictive of posttherapy distant China. Li WF.M. outcomes after radiotherapy alone for patients with early-stage nasopha- 17. et al. Ma J.69(1):239–241.62(3): We thank the Good Clinical Practices Center and all of the colleagues and 672–679. Pajak TF. Guangzhou. C-NQ. 41.oxfordjournals.-Q. Toxicity criteria of the Radiation Therapy concurrent chemoradiotherapy in the patients with locally advanced naso- Oncology Group (RTOG) and the European Organization for Research pharyngeal cancer. et al. carcinoma. 1982. 28.68(10):2120–2124.40(2):62–70. Ma J.-Y. et al.-Q. Mendenhall 21. J Clin Oncol. Huang YJ. Cancer. Guangzhou. Lee MA. Kim TH. Chan AT. Issue 23  |  December 7. Campbell MJ. Xiao WW. 2005. Kim KH. Cancer. Biometrika. Chen L. the writing of the terns. Zhao C. patients who were involved in this trial. Int J Radiat Oncol Biol Phys. Treatment carcinoma in China. Non-parametric estimation from incomplete obser. How does intensity-modulated radio- 35(1):1–39. et al. NY: Springer-Verlag. Multicenter phase II trial of mitoxan- 18. Asia: an Asian-Oceanian Clinical Oncology Association Group study.). M-YC. Choo R.-Q. Wu HG. Au GK. 2004. Kaplan EL. To prove the rationality of ’92 staging trone in patients with advanced nasopharyngeal carcinoma in Southeast with results of nasopharyngeal carcinoma treated by stratified multitherapies. Tailoring distant metastatic imaging for 31(5):1341–1346. therapy versus conventional two-dimensional radiotherapy influence the 26. 2003. Treatment results of 1070 patients for statistical assistance. 2nd ed. Y-QX. Tang Y. Design and analysis of randomized clinical 3684–3690.58(3):688–693. Concomitant weekly cisplatin and altered fractionation radiotherapy ing faciocervical portals at isocenter for radiotherapy of nasopharyngeal in locally advanced head and neck cancer. Song CH.org/ at Univ of Iowa-Law Library on March 14. Cheng NN. 16. Loh KS. 2015 cancer treatment. Cox DR. Newlin HE. Zhao C. Lai SZ.47(1):207–214. Zong J. L. 103. Riggs CE.M. Machin D. Histological Typing of Tumors of the Upper ryngeal carcinoma. Greene FL. Liu contributed equally to this article. et al. with or without chemotherapy for nasopharyngeal carcinoma: radiation vations. Cui NJ. 2009. Trotti A. 2011. Cao XP. A probe of design methods of block shield. Bulletin Chin Cancer. Chin J Cancer. 1995. 7th ed. Byrd DR. 1993.34(2): treatment results in nasopharyngeal carcinoma patients? Int J Radiat Oncol 187–220. 1958. 2007. 1994. The authors have therapy alone for patients with stage I-II nasopharyngeal carcinoma. Lu JJ.). of China and Department of Nasopharyngeal Carcinoma. 19.). 2010. Leung TW. Guo. LG.19(5):460–462. J Am Stat Assoc. Chemotherapy for recurrent or metastatic carcinoma Studies. 31. Staquet M. et al. Armitage P. Sham JS. Lu TX. or the decision to submit the article for publication. Ngai A. H-YM. Treatment outcome of cisplatin-based 22. Stetz J. 2008. et al. J Clin Oncol. J R Stat Soc B. Kirwan JM. et al. Kwong DL. Chin J Clin Pharmacol Ther. H-QM). treatment (NDST).74(4):1070–1076. and interpretation of the data. People’s Republic failure in patients with early-stage nasopharyngeal carcinoma of undiffer. Schoenfeld D. Y. 2000. Zee B.). 38. Q.27(10):620–622. Min H. M-QD. K-JC. UK: Blackwell. Br J Cancer. Chin J Oncol. therapy oncology group phase II trial 0225. Fritz AG. Miller AB. N-WL. Winkler A. People’s Republic of China (M-HH). Intensity-modulated radiation therapy 24. Chua DT. 2005.116(19):4533–4540.98(1):74–80. the Sun Yat-sen University Clinical therapy alone.-Q. 1977.19(10):930–933. Hong MH. the Sci-Tech Project Foundation therapy followed by radiotherapy in early-stage nasopharyngeal carci.). Sun Yat-sen entiated type. Fayers PM. The funding agencies had no role in the design of the with nasopharyngeal carcinoma: an analysis of survival and failure pat. Regression models and life tables. Kumar MB.14(8):538–542. Respiratory Tract and Ear. trials requiring prolonged observation of each patient.118(4):663–670. Introduction of software of the new drug statistical ence. Good Clinical Practices Center. Foundation of China (81072226 to H. Shanmugarantnam K. et al. Heo DS. 1981. Guangdong Provincial Medical Research Foundation (B2006063 30. analysis.