Preliminary Results of a Randomized Study on Therapeutic Gain by Concurrent Chemotherapy for Regionally-Advanced Nasopharyngeal Carcinoma: N
http://www.100md.com
《临床肿瘤学》
the Departments of Clinical Oncology, Pamela Youde Nethersole Eastern Hospital, Queen Elizabeth Hospital, Tuen Mun Hospital, and Queen Mary Hospital, Hong Kong, China
Department of Biostatistics, University of Wisconsin Medical School, Madison, WI
Comprehensive Cancer Trials Unit, Chinese University of Hong Kong, China
Ontario Cancer Institute, Princess Margaret Hospital, Toronto, Canada
ABSTRACT
PURPOSE: This randomized study compared the results achieved by concurrent chemoradiotherapy (CRT) versus radiotherapy (RT) alone for nasopharyngeal carcinoma (NPC) with advanced nodal disease.
PATIENTS AND METHODS: Patients with nonkeratinizing/undifferentiated NPC staged T1-4N2-3M0 were randomized to CRT or RT. Both arms were treated with the same RT technique and dose fractionation. The CRT patients were given cisplatin 100 mg/m2 on days 1, 22, and 43, followed by cisplatin 80 mg/m2 and fluorouracil 1,000 mg/m2/d for 96 hours starting on days 71, 99, and 127.
RESULTS: From 1999 to January 2004, 348 eligible patients were randomly assigned; the median follow-up was 2.3 years. The two arms were well-balanced in all prognostic factors and RT parameters. The CRT arm achieved significantly higher failure-free survival (72% v 62% at 3-year, P = .027), mostly as a result of an improvement in locoregional control (92% v 82%, P = .005). However, distant control did not improve significantly (76% v 73%, P = .47), and the overall survival rates were almost identical (78% v 78%, P = .97). In addition, the CRT arm had significantly more acute toxicities (84% v 53%, P < .001) and late toxicities (28% v 13% at 3-year, P = .024).
CONCLUSION: Preliminary results confirmed that CRT could significantly improve tumor control, particularly at locoregional sites. However, there was significant increase in the risk of toxicities and no early gain in overall survival. Longer follow-up is needed to confirm the ultimate therapeutic ratio.
INTRODUCTION
Nasopharyngeal carcinoma (NPC) is different from other head and neck cancers because of its unique epidemiology, natural behavior, and therapeutic considerations. The distribution pattern is peculiarly skewed with age-standardized incidence rate (per 100,000 male population) ranging from more than 20 in Southeast Asian populations to 0.5 in predominantly white populations.1 There is also marked variation in histologic pattern among different ethnic groups, with correlated differences in responsiveness to radiotherapy (RT) and survival.2
Although NPC is a radiosensitive tumor, results of conventional RT technique and fractionation are unsatisfactory for patients with advanced locoregional disease. The Intergroup 0099 Study (IGS)3,4 comparing concurrent chemoradiotherapy (CRT) versus RT alone, for patients with stage III-IVB disease by the American Joint Committee on Cancer Staging System, fourth edition (AJCC-4),5 was an important milestone. Not only was this the first trial with significant benefit in overall survival (OS), but also the magnitude of gain was so impressive (78% v 47% at 3-year),3 that this regimen became widely recommended as the standard treatment for patients with advanced NPC.
However, several serious concerns remain. First, the true magnitude of benefit might be smaller, as the result of the RT arm in the IGS was exceptionally poor. Other series on patients with similar stages showed that 3-year OS 70% could be achieved by RT alone.6-8 A retrospective analysis by Chua et al9 showed that patients treated with the IGS regimen had almost the same OS and distant control as a matched cohort treated by RT alone (P > .84). The efficacy of CRT for distant control is particularly uncertain because two studies with excellent locoregional control ( 93%) using new RT technologies,10,11 still had high incidence of distant failure ( 25%) despite extensive use of the IGS regimen ( 75% of patients). Second, the ultimate impact on therapeutic ratio is unknown, as data on late toxicities were lacking. Third, because the proportion of patients with keratinizing squamous cell carcinoma was high ( 22%) in the IGS,3 the applicability of its results to Asian patients is uncertain. Fourth, the proportions of patients who could complete the scheduled concurrent and adjuvant chemotherapy in the IGS were only 63% and 55%, respectively3; the compliance and tolerance in Asian patients are worrisome.
Confirmatory trials were hence conducted by the Hong Kong Nasopharyngeal Cancer Study Group to resolve these important issues. To search for the most cost-effective treatment strategy for different risk groups, patients with stages III-IVB were segregated into two groups and studied in two parallel trials. The NPC-9901 Trial on patients with T1-4N2-3M0 disease (whose key problem is distant failure) was designed to confirm the therapeutic ratio achieved by the IGS regimen; while the NPC-9902 Trial on patients with T3-4N0-1M0 disease (whose key problem is local failure) was designed not only to confirm the therapeutic ratio achieved by the IGS regimen, but also to compare that achieved by accelerated fractionation. The present article is the preliminary report of the NPC-9901 Trial.
PATIENTS AND METHODS
The trial was approved by the institutional ethics committees of individual participating center, and monitored by an independent data monitoring committee composed of radiation oncologist, medical oncologist, and statistical consultants.
Eligibility Criteria
Patients who fulfilled all the following criteria were eligible for this study:
Biopsy-proven nonkeratinizing (type 2.1) or undifferentiated (type 2.2) carcinoma of the nasopharynx by the WHO classification12;
N2 or N3 disease by the staging system of the AJCC and the International Union Against Cancer, fifth edition (AJCC/UICC-5)13,14;
No gross evidence of distant metastasis;
Adequate hematologic function: total leukocyte count (WBC) 4,000/μL and platelet 100,000/μL;
Adequate renal function: creatinine clearance (CrCl) 60 mL/min;
Satisfactory performance status: 2 by the Eastern Cooperative Oncology Group system.
Exclusion criteria included keratinizing squamous cell carcinoma or adenocarcinoma, age 70 years, pregnancy or lactation, history of previous treatment, or prior malignancy (except adequately treated carcinoma-in-situ of the cervix or basal/squamous cell carcinoma of the skin).
Pretreatment Evaluation and Randomization
All patients were evaluated by complete physical examination, fiberoptic nasopharyngoscopy, computed tomography or magnetic resonance imaging (MRI) of the nasopharyngeal region, chest x-ray, complete blood count, renal and liver function tests, and lactate hydrogenate (LDH). Additional investigations were performed only for those with suspicious findings or abnormal biochemical profile.
All patients were required to give written informed consent before registration. They were stratified by treatment center, T-category (T1-2 v T3–4), and N-category (N2 v N3). The blocked randomization scheme15 was used to allocate patients in equal proportions to the two treatment arms.
Radiotherapy
Patients in both arms were irradiated with megavoltage photons using the same RT technique and dose schedules in line with the treatment policy adopted by each center. All were treated with conventional fractionation of 2 Gy per fraction, 5 daily fractions per week, to a total dose 66 Gy.
Technique ranged from conventional two-dimensional technique to three-dimensional conformal or intensity-modulated technique throughout. All potential sites of local infiltration and bilateral cervical lymphatics were irradiated to 50 Gy. Additional boosts to the parapharyngeal space, the primary or nodal sites could be given (when indicated), the boost field was confined to the involved site with exclusion of critical structures, and the boost dose did not exceed 20 Gy.
Chemotherapy
Patients randomly assigned to the CRT arm were given additional chemotherapy using the IGS regimen. Cisplatin (CDDP) 100 mg/m2 was given intravenously every 3 weeks on days 1, 22, and 43 in concurrence with RT. Subsequent adjuvant chemotherapy using a combination of CDDP 80 mg/m2 intravenously and fluorouracil 1,000 mg/m2/d by 96-hour infusion was given every 4 weeks starting on days 71, 99, and 127.
Dose modifications were based on the nadir blood counts and interim toxicities of the preceding cycle. Chemotherapy was withheld until WBC 3,500/μL and platelet 100,000/μL. If the nadir of WBC was 2,500/μL and platelet 75,000/μL, full scheduled doses were given. If the nadir of WBC was 1,500/μL (less than 2,500/μL) or platelet 50,000/μL (less than 75,000/μL), CDDP was decreased by 20 mg/m2. If the nadir of WBC was less than 1,500/μL or platelet less than 50,000/μL, CDDP was decreased by 40 mg/m2 and fluorouracil by 200 mg/m2.
For renal toxicity, CDDP was withheld until CrCl 60 mL/min. If the nadir of CrCl was 40 (less than 60 mL/min), CDDP was decreased by 20 mg/m2 or replaced by carboplatin (using area under the curve of five). If the nadir of CrCl was less than 40 mL/min, or the patient developed otologic/neurologic toxicity grade 3, CDDP was stopped totally.
The dose of fluorouracil was decreased by 200 mg/m2 for grade 2, and 400 mg/m2 for grade 3 stomatitis or gastrointestinal toxicities, and stopped permanently if the patient developed drug-related angina.
Patient Assessments and Follow-Up
Acute RT-related toxicities were graded according to the Acute Radiation Morbidity Scoring Criteria of the Radiation Therapy Oncology Group16 and chemotherapy-related toxicities by the WHO criteria.17 Incidents of toxicities grade 3 (except nausea or alopecia) were recorded.
The first assessment of tumor response was performed 6 weeks to 16 weeks after completion of RT. All patients were assessed by complete physical examination and fiberoptic nasopharyngoscopy. Further investigations with computed tomography/MRI or other tests were arranged when indicated. Management of residual disease, if found, was decided by individual center. For statistical purposes, persistent primary/nodal disease at 16 weeks after completion of RT was taken as locoregional failure.
Patients were followed-up at least every 3 months during the first 3 years, and then every 6 months thereafter until death. The earliest dates of detecting tumor relapse at different sites were recorded. Salvage treatment was attempted, as far as possible, in line with the policies of individual centers. Late toxicities (occurring > 90 days after commencement of RT) were defined and graded according to the Late Morbidity Scoring Criteria of Radiation Therapy Oncology Group.16 The earliest date of detecting toxicity (except xerostomia and dental caries) grade 3 was recorded.
Statistical Methods
The primary end points of this study include failure-free survival (FFS), defined as the time to first failure at any site, and progression-free survival (PFS), defined as the time to first failure or death as a result of any cause. Assuming that the 5-year FFS by RT alone was 40%, the target accrual for detecting a 15% difference with 80% power was 340 patients.
Secondary end points for treatment efficacy included: locoregional failure-free rate (LR-FFR; persistence/recurrence in the nasopharyngeal and/or cervical region), distant failure-free rate (D-FFR; hematogenous metastasis), and overall survival (OS; death as a result of any cause).
Secondary end points for major treatment toxicities (grade 3) included crude incidence rates and the actuarial rate of late toxicities. For patients who had re-irradiation for treatment of locoregional relapses, events were censored at commencement of re-irradiation for assessing toxicities incurred solely by the primary treatment.
All events were measured from the date of registration, and the analyses were performed on an intention-to-treat basis. The actuarial rates were calculated by the product-limit method,18 and the differences compared by the Mantel-Haenszel log-rank test.19 The Breslow-Day 2 test was used for comparing incidence rates and categoric variables, and Student's t test was used for comparing the means of continuous variables. Multivariate analyses by Cox proportional hazards regression model20 were performed to test the independent significance of different factors. Covariates include host factors (ie, age, sex, performance status), tumor factors (ie, stage, LDH), RT (ie, total dose, overall treatment time, boost), and chemotherapy intervention (ie, treatment arm). All statistical tests were two-sided, and P values less than .05 were considered statistically significant. Our protocol specified two analyses to be performed, one at completion of accrual and another with complete 5-year follow-up.
RESULTS
Patient Characteristics
March 1999 to January 2004, 348 eligible patients were randomly assigned. Participating centers included Pamela Youde Nethersole Eastern Hospital, Tuen Mun Hospital, Queen Mary Hospital and Queen Elizabeth Hospital from Hong Kong, and Princess Margaret Hospital from Canada.
The two arms were well-balanced in all patient characteristics and tumor factors (Table 1). Ninety-five percent of patients had been regularly followed-up until death or the latest scheduled assessment. The median duration of observation was 2.3 years (range, 0.2 to 5.3).
Treatment and Compliance
The two arms were well-balanced in all RT parameters (Table 1). The techniques used were balanced (P = .87). The median total dose was 68 Gy, and overall time 46 days in both arms. Parapharyngeal boost via a postero-lateral field was given to 29% in the CRT arm and 28% in the RT arm (P = .89). Among those given additional boosts, the median dose given was 10 Gy. Only two patients in the CRT arm and one patient in the RT arm failed to complete the scheduled total dose; and only one patient in each arm had prolongation more than 7 days.
Four patients had major protocol violations: two patients in the CRT arm had no chemotherapy because of an incidental cause and the patient's refusal, and two patients in the RT arm were given chemotherapy as a result of rapid disease progression.
In the CRT arm, 65% of patients completed all six cycles of chemotherapy, and 79% had five cycles. The mean number of cycles given during the concurrent phase was 2.5 (± 0.6) and 2.6 (± 1.3) cycles during the adjuvant phase. Table 2 shows the details of compliance during each phase. The mean total dose of CDDP and fluorouracil given were 444 and 9,145 mg/m2 (82% and 76% of the total scheduled doses, respectively).
Acute Toxicities
Altogether 143 patients (84%) in the CRT arm had one or more toxicities, as compared to 93 patients (53%) in the RT arm (Table 3; P < .001). The majority of toxicities were uneventful and patients recovered rapidly, but the CRT arm had more grade 4 toxicities (12% v 1%) and one patient succumbed.
In addition to chemotherapy-related toxicities, the CRT arm also had significantly higher incidence of RT-related mucositis (62% v 48%, P = .01), but not of skin reaction (20% v 17%, P = .41).
The total incidence of chemotherapy-related toxicities in the CRT arm was 59%. The most common were leukopenia and/or neutropenia (32%); one patient died of neutropenic sepsis. The next most common chemotherapy-related toxicities included anemia (20%) and vomiting (19%). Besides the well-known toxicities, 2% had reactivation of hepatitis and 1% pulmonary tuberculosis.
Tumor Control
Only 2% in the CRT arm and 1% in the RT arm failed to achieve complete response at 16 weeks after completion of RT. At the time of analyses (September 2004), 39 patients (23%) in the CRT arm and 58 patients (33%) in the RT arm had failed at one or more sites (Table 4). The FFS showed significant improvement (Fig 1; 72% v 62% at 3-year, P = .027). The hazard ratio (HR) was 0.60; 95% CI, 0.40 to 0.92.
The CRT arm achieved significantly higher LR-FFR (Fig 2A; 92% v 82%, P = .005; HR = 0.28; 95% CI, 0.12 to 0.64). However, the improvement in D-FFR was insignificant (Fig 2B; 76% v 73%, P = .47; HR = 0.84; 95% CI, 0.53 to 1.34).
Among the patients with relapse, the proportion retreated aggressively was not significantly different: 59% (23 out of 39) in the CRT arm v 66% (38 out of 58) in the RT arm (P = 0.51). Salvage treatment given included chemotherapy (36% v 47%), re-irradiation (8% v 12%), and surgery (3% v 14%). The resultant salvage rate was higher in the RT arm, but the difference was insignificant (8% v 19%, P = .12).
Hence, the incidence of death as a result of disease progression (Table 4) was only slightly lower in the CRT arm (15% v 19%). However, there were extra deaths as a result of other causes (4.7% v 1.2%) in the CRT arm, including 1.1% treatment mortality, 1.1% unknown causes, and 3.5% incidental causes (ie, infection, cerebral vascular accident, suicide, and second malignancy), while only 1.1% in the RT arm died of incidental causes (ie, infection, dermatomyositis). The 3-year OS was 78% in both arms (Fig 3; P = .97; HR = 0.99; 95% CI, 0.62 to 1.58).
Altogether 46 patients (27%) in the CRT arm and 60 patients (34%) in the RT arm had experienced treatment failure or had died. Though the 3-year PFS was higher in the CRT arm (70% v 61%, P = .10), the trend of improvement was short of statistical significance (HR = 0.70; 95% CI, 0.47 to 1.04).
Subset analyses (Table 5) showed that the effect on D-FFR and OS was insignificant in all subsets. The greatest magnitude of differences was achieved among the 142 patients with stage IV disease (D-FFR, 62% v 52%; OS, 66% v 61%), but none of the end points reached statistical significance in this group.
On multivariate analyses (Table 6), CRT was an independent significant factor not only for LR-FFR and FFS, but also for PFS with hazard for progression being 0.59 (95% CI, 0.40 to 0.89; P = .012). However, the impact on OS remained insignificant with hazard for death being 0.79 (95% CI, 0.48 to 1.29; P = .34).
Among the other prognostic factors, total RT dose was significant for all end points including OS with hazard for death being 0.94 (95% CI, 0.92 to 0.97) per Gy increase. Both stage and LDH were significant for all end points except LR-FFR.
Late Toxicities
Thirty-four patients (20%) in the CRT arm and 20 patients (11%) in the RT arm developed one or more late toxicities attributed to the primary course of treatment (Table 7). The 3-year actuarial rate of late toxicities was higher in the CRT arm (Fig 4; 28% v 13%, P = .024; HR = 1.87; 95% CI, 1.08 to 3.26). This was mostly because of increased incidence of otologic toxicities (14% v 8%), peripheral neuropathy (2% v 0%), and endocrine dysfunction (4% v 1%). The majority of toxicities were grade 3 in severity. Only 3% in the CRT arm and 1% in the RT arm had grade 4 toxicities. Damages to neurologic structures were rare, but one patient (0.6%) in the CRT arm developed last 4 cranial nerves palsy and died of aspiration pneumonia.
DISCUSSION
Concurrent CRT is one of the most promising strategies for improving tumor control of advanced NPC, as this has potential for both enhancing the local effect of RT and eradicating micrometastases. There is little controversy that this is more potent than sequential chemotherapy, and is, thus far, the only strategy with significant survival benefit proven by randomized trials. However, with rapid advances in RT technologies, it is important that we provide patients with accurate information about the magnitude of benefit and the balance against excessive toxicities. The current trial, focused on patients with N2-3 disease, provides a good opportunity for studying the efficacy of the IGS regimen particularly for distant control. The findings are valid as the two arms were well-balanced in all prognostic factors and RT parameters (Table 1). Although the lack of routine imaging studies for exclusion of initial distant metastases and subsequent monitoring might lead to under-diagnosis of failures, this would not bias the current conclusions, as this concern applies to both treatment arms. No trends of potential bias caused by the bad news travels fast phenomenon21 are found in our data on additional testing. Furthermore, the sample size in the current study is large compared with most other trials.3,7,22,23 The only limitation is that the follow-up period (median, 2.3 years) is still relatively short, particularly for assessing late toxicities. Thus, the current conclusions should be taken as preliminary.
Table 8 summarizes our results together with other randomized trials on concurrent CRT3,7,22-24 to give a more comprehensive perspective of our current knowledge. It should be noted that there are variations in staging system, patients included in previous trials3,7,22,24 were actually staged II–IVB by the criteria of the current staging system,13,14 while the current trial only included the regionally-advanced group within stages III–IVB. Cross-series comparison is difficult.
The latest data published and/or presented at academic conferences are quoted to show the most updated data. In addition to the progress report by the Intergroup,4 a new trial presented by Wee et al23 also showed significant survival benefit by the IGS regimen (2-year OS, 85% v 77%, P = .02). However, it must be cautioned that the results currently listed are based on their abstracts published in the conference proceedings; full publications with thorough peer review are still pending.
The most recent publication is the progress report on the large series of 350 patients by Chan et al,25 comparing RT versus CRT using CDDP 40 mg/m2 weekly during the concurrent phase and no adjuvant therapy. With a median follow-up of 5.5 years, both FFS (the primary end point in their first report)24 and OS (the primary end point in their progress report)25 showed a positive trend short of statistical significance; the 5-year results for FFS were 60% v 52% (P = .16) and for OS were 70% v 59% (P = .065). Only in the Cox regression analysis, after adjusting for T-stage, overall stage, and age, did the difference in OS reach statistical significance (HR = 0.74; 95% CI, 0.5 to 1.0; P = .049).
Our data supports the conclusion by IGS3,4 and Lin et al7 that CRT could significantly improve locoregional control (Fig 2A; 92% v 82% at 3-years, P = .005). In concurrence with Lin et al, our primary end point FFS also achieved significant improvement (Fig 1; 72% v 62%, P = .027). The magnitude of difference in 3-year PFS (71% v 60%, P = .10) was substantially smaller than the original IGS report (69% v 24%),3 though multivariate analyses did show that CRT intervention was a significant factor for PFS (Table 6; HR = 0.59; 95% CI, 0.40 to 0.89; P = .012).
However, our data concur with the finding by Chan et al24,25 that distant control did not show significant improvement (Fig 2B; 76% v 73%, P = .47). Furthermore, with aggressive salvage after relapse, the OS were almost identical up to 3 years (Fig 3; 78% v 78%, P = .97). This phenomenon occurred in all subsets (Table 5). Though the 142 patients with stage IV disease did show greater magnitude of benefit in D-FFR (62% v 52%) and OS (66% v 61%), the differences did not reach statistical significance (P > .23).
Closer review of the trials with 5-year results for patients with stages II–IVB (by AJCC/UICC-5) showed that absolute survival gain achieved was 30% (67% v 37%, P < .01) by IGS,4 18% (72% v 54%, P = .002) by Lin et al,7 and 11% (70% v 59%, P = .065) by Chan et al.25 It is apparent that the exceptionally great magnitude of survival gain by IGS is mostly because of the poor result of the RT arm. One possible explanation is the difference in histologic pattern (28% of keratinizing type in the IGS as compared with < 4% in other series).
Our current failure to achieve significant survival benefit is probably not caused by suboptimal chemotherapy, because the dose of CDDP scheduled in our implementation of the IGS regimen was the highest among the different regimens used, and our compliance was not exceptionally poor (65% completed all six cycles, and the mean total dose of CDDP given was 444 mg/m2). Although CRT did cause significant increases in incidence and severity of toxicities both acute (Table 3; 84% v 53%, P < .001) and late (Table 5; Fig 4; 28% v 13% at 3-year, P = .024), the majority of toxicities were uneventful grade 3 toxicities. The IGS regimen could be accepted as reasonably well-tolerated even for Asian patients, but it should be noted that in addition to 1% treatment mortality (because of neutropenic sepsis and aspiration pneumonia), the CRT had an extra 4% mortality caused by unknown and incidental causes (including infection, cerebral vascular accident, and suicide).
Given the higher proportion of advanced cases in the current series (all being stages III–IVB by AJCC/UICC-5), it appears that the OS achieved by our CRT arm was not inferior to other series, but it was the favorable results of the RT arm that narrowed any potential gain. The obvious question is whether the results of the current RT arm are reproducible. To obtain the most representative data on treatment results in the modern era, the Hong Kong Nasopharyngeal Cancer Study Group has conducted a retrospective study of patients treated in all local public centers during 1996 to 2000.26 The technological provision was slightly inferior as only 32% of the series was staged by MRI and 10% irradiated by three-dimensional conformal or intensity-modulated technique (compared with 69% and 50%, respectively, in the current series). The 3-year OS for the 905 consecutive patients with stages III–IVB disease (by AJCC/UICC-5) treated by RT alone was 74%. Hence, the results achieved in the current trial should be reproducible, especially for nonkeratinizing/undifferentiated carcinoma. Indeed, even better results might be achievable, at least for locoregional control, with increasing availability of modern technologies.
In conclusion, our preliminary results confirm that CRT using the IGS regimen could improve tumor control, particularly at locoregional sites. It is possible that corresponding gain in survival may manifest with longer follow-up. We support recommending this option to patients with advanced NPC. However, patients should be duly informed that with improving RT technologies, the differential gain in survival might be small and there is risk of significant increase in toxicities. More potent therapy for distant control and more accurate prognostication to minimize unnecessary over-treatment are urgently needed. Estimation of sample size for future trials should be based on more realistic magnitude of benefit and higher baseline results by modern RT to avoid under-power.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank the data monitoring committee (K. Ang, P. Johnson, R. Chappell, and J. van Houwelingen), data manager (M.W. Chan), internal monitors (W.M. Ling, W. Sung, C.M. Lam, and J Yuen), et al of participating centers for their contribution.
NOTES
Supported by grants from the Hong Kong Cancer Fund, Ho Hung Chiu Medical Foundation Limited, and the Hong Kong Anti-Cancer Society, Hong Kong.
Presented in part at the 40th Annual Meeting of the American Society of Clinical Oncology, New Orleans, LA, June 5-8, 2004), and the 46th Annual Meeting of the American Society of Therapy Radiation Oncology, Atlanta, GA, October 3-7, 2004.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Parkin DM, Whelan SL, Ferlay J, et al: Cancer Incidence in Five Continents. IARC Scientific Publications Number 155, Vol. 8. Lyon, France, IARC Press, 2002, pp 534-536
Marks JE, Phillips JL, Menck HR: The National Cancer Data Base Report on the relationship of race and national origin to the histology of nasopharyngeal carcinoma. Cancer 83:582-588, 1998
Al-Sarraf M, LeBlanc M, Giri PG, et al: Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: Phase III randomized Intergroup study 0099. J Clin Oncol 16(4):1310-1317, 1998
Al-Sarraf M, LeBlanc M, Giri PGS, et al: Chemo-radiotherapy (CT-RT) vs radiotherapy (RT) in patients (PTS) with advanced nasopharyngeal cancer (NPC). Intergroup (0099) (SWOG 8892, RTOG 8817, ECOG 2388) Phase III Study: Progress report. J Clin Oncol 17:385a, 1998 (abstr 1483)
Beahrs O, Henson DE, Hutter RVP, et al: American Joint Committee on Cancer: Manual for Staging of Cancer, Fourth Edition. Philadelphia, PA, Lippincott, 1992, pp 33-38
Chow E, Payne D, O'Sullivan B, et al: Radiotherapy alone in patients with advanced nasopharyngeal cancer: Comparison with an intergroup study –Is combined modality treatment really necessary? Radiother Oncol 63:269-274, 2002
Lin JC, Jan JS, Hsu CY, et al: Concurrent chemoradiotherapy versus radiotherapy alone for advanced nasopharyngeal carcinoma: Positive effect on overall and progression-free survival. J Clin Oncol 21:631-637, 2003
Wolden SL, Zelefsky MJ, Kraus DH, et al: Accelerated concomitant boost radiotherapy and chemotherapy for advanced nasopharyngeal carcinoma. J Clin Oncol 19:1105-1110, 2001
Chua DTT, Sham J, Au G, et al: Concomitant chemoirradiation for Stage III-IV nasopharyngeal carcinoma in Chinese patients: Results of a matched cohort analysis. Int J Radiat Oncol Biol Phys 53:334-343, 2002
Le QT, Tate D, Koong A, et al: Improved local control with stereotactic radiosurgical boost in patients with nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 56:1046-1054, 2003
Lee N, Xia P, Quivey JM, et al: Intensity-modulated radiotherapy in the treatment of nasopharyngeal carcinoma: An update of the UCSF experience. Int J Radiat Oncol Biol Phys 53:12-22, 2002
Shanmugarantnam K, Sobin LH. Histological Typing of Tumors of the Upper Respiratory Tract and Ear, Second Edition. New York, NY, Springer-Verlag, 1991, pp 32-33
Fleming ID, Cooper JS, Henson DE, et al: American Joint Committee on Cancer: AJCC Cancer Staging Manual, Fifth Edition. Philadelphia, PA, Lippincott-Raven, 1997, pp 31-39
Sobin LH, Wittekind Ch. International Union Against Cancer (UICC): TNM Classification of Malignant Tumors, Fifth Edition. New York, NY, Wiley-Liss, 1997, pp 25-32
Freedman J, Furberg C, DeMets D. Fundamentals of Clinical Trials. New York, NY, Springer-Verlag, 1998
Cox JD, Stetz J, Pajak TF: Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 31:1341-1346, 1995
Miller AB, Hoogstraten B, Staquet M, et al: Reporting results of cancer treatment. Cancer 47:207-214, 1981
Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Peto R, Pike MC, Armitage P, et al: Design and analysis of randomized clinical trials requiring prolonged observation of each patient. Br J Cancer 35:1-39, 1977
Cox DR: Regression models and life tables. J Roy Stat Soc B 34:187-220, 1972
Schiffer CA, Dodge R, Larson RA: Long-term follow-up of Cancer and Leukemia Group B studies in acute myeloid leukemia. Cancer 80:2210-2214, 1997
Kwong DLW, Sham JST, Au GKH, et al: Concurrent and adjuvant chemotherapy for nasopharyngeal carcinoma: A factorial study. J Clin Oncol 22:2643-2653, 2004
Wee J, Tan EH, Tai BC, et al: Phase III randomized trial of radiotherapy versus concurrent chemoradiotherapy followed by adjuvant chemotherapy in patients with AJCC/UICC (1997) stage 3 and 4 nasopharyngeal cancer of the endemic variety. J Clin Oncol 23:487, 2004 (abstr 5500)
Chan ATC, Teo PML, Ngan RK, et al: Concurrent chemotherapy-radiotherapy compared with radiotherapy alone in locoregionally advanced nasopharyngeal carcinoma: Progression-free survival analysis of a phase III randomized trial. J Clin Oncol 20:2038-2044, 2002
Chan AT, Leung SF, Ngan RK, et al: Overall survival after concurrent cisplatin-radiotherapy compared with radiotherapy alone in locoregionally advanced nasopharyngeal carcinoma. J Natl Cancer Inst 97(7):536-539, 2005
Lee AW, Sze WM, Au JS, et al: Treatment results for nasopharyngeal carcinoma in the modern era: The Hong Kong experience. Int J Radiat Oncol Biol Phys 61(4):1107-1116, 2005(Anne W.M. Lee, W.H. Lau, )
Department of Biostatistics, University of Wisconsin Medical School, Madison, WI
Comprehensive Cancer Trials Unit, Chinese University of Hong Kong, China
Ontario Cancer Institute, Princess Margaret Hospital, Toronto, Canada
ABSTRACT
PURPOSE: This randomized study compared the results achieved by concurrent chemoradiotherapy (CRT) versus radiotherapy (RT) alone for nasopharyngeal carcinoma (NPC) with advanced nodal disease.
PATIENTS AND METHODS: Patients with nonkeratinizing/undifferentiated NPC staged T1-4N2-3M0 were randomized to CRT or RT. Both arms were treated with the same RT technique and dose fractionation. The CRT patients were given cisplatin 100 mg/m2 on days 1, 22, and 43, followed by cisplatin 80 mg/m2 and fluorouracil 1,000 mg/m2/d for 96 hours starting on days 71, 99, and 127.
RESULTS: From 1999 to January 2004, 348 eligible patients were randomly assigned; the median follow-up was 2.3 years. The two arms were well-balanced in all prognostic factors and RT parameters. The CRT arm achieved significantly higher failure-free survival (72% v 62% at 3-year, P = .027), mostly as a result of an improvement in locoregional control (92% v 82%, P = .005). However, distant control did not improve significantly (76% v 73%, P = .47), and the overall survival rates were almost identical (78% v 78%, P = .97). In addition, the CRT arm had significantly more acute toxicities (84% v 53%, P < .001) and late toxicities (28% v 13% at 3-year, P = .024).
CONCLUSION: Preliminary results confirmed that CRT could significantly improve tumor control, particularly at locoregional sites. However, there was significant increase in the risk of toxicities and no early gain in overall survival. Longer follow-up is needed to confirm the ultimate therapeutic ratio.
INTRODUCTION
Nasopharyngeal carcinoma (NPC) is different from other head and neck cancers because of its unique epidemiology, natural behavior, and therapeutic considerations. The distribution pattern is peculiarly skewed with age-standardized incidence rate (per 100,000 male population) ranging from more than 20 in Southeast Asian populations to 0.5 in predominantly white populations.1 There is also marked variation in histologic pattern among different ethnic groups, with correlated differences in responsiveness to radiotherapy (RT) and survival.2
Although NPC is a radiosensitive tumor, results of conventional RT technique and fractionation are unsatisfactory for patients with advanced locoregional disease. The Intergroup 0099 Study (IGS)3,4 comparing concurrent chemoradiotherapy (CRT) versus RT alone, for patients with stage III-IVB disease by the American Joint Committee on Cancer Staging System, fourth edition (AJCC-4),5 was an important milestone. Not only was this the first trial with significant benefit in overall survival (OS), but also the magnitude of gain was so impressive (78% v 47% at 3-year),3 that this regimen became widely recommended as the standard treatment for patients with advanced NPC.
However, several serious concerns remain. First, the true magnitude of benefit might be smaller, as the result of the RT arm in the IGS was exceptionally poor. Other series on patients with similar stages showed that 3-year OS 70% could be achieved by RT alone.6-8 A retrospective analysis by Chua et al9 showed that patients treated with the IGS regimen had almost the same OS and distant control as a matched cohort treated by RT alone (P > .84). The efficacy of CRT for distant control is particularly uncertain because two studies with excellent locoregional control ( 93%) using new RT technologies,10,11 still had high incidence of distant failure ( 25%) despite extensive use of the IGS regimen ( 75% of patients). Second, the ultimate impact on therapeutic ratio is unknown, as data on late toxicities were lacking. Third, because the proportion of patients with keratinizing squamous cell carcinoma was high ( 22%) in the IGS,3 the applicability of its results to Asian patients is uncertain. Fourth, the proportions of patients who could complete the scheduled concurrent and adjuvant chemotherapy in the IGS were only 63% and 55%, respectively3; the compliance and tolerance in Asian patients are worrisome.
Confirmatory trials were hence conducted by the Hong Kong Nasopharyngeal Cancer Study Group to resolve these important issues. To search for the most cost-effective treatment strategy for different risk groups, patients with stages III-IVB were segregated into two groups and studied in two parallel trials. The NPC-9901 Trial on patients with T1-4N2-3M0 disease (whose key problem is distant failure) was designed to confirm the therapeutic ratio achieved by the IGS regimen; while the NPC-9902 Trial on patients with T3-4N0-1M0 disease (whose key problem is local failure) was designed not only to confirm the therapeutic ratio achieved by the IGS regimen, but also to compare that achieved by accelerated fractionation. The present article is the preliminary report of the NPC-9901 Trial.
PATIENTS AND METHODS
The trial was approved by the institutional ethics committees of individual participating center, and monitored by an independent data monitoring committee composed of radiation oncologist, medical oncologist, and statistical consultants.
Eligibility Criteria
Patients who fulfilled all the following criteria were eligible for this study:
Biopsy-proven nonkeratinizing (type 2.1) or undifferentiated (type 2.2) carcinoma of the nasopharynx by the WHO classification12;
N2 or N3 disease by the staging system of the AJCC and the International Union Against Cancer, fifth edition (AJCC/UICC-5)13,14;
No gross evidence of distant metastasis;
Adequate hematologic function: total leukocyte count (WBC) 4,000/μL and platelet 100,000/μL;
Adequate renal function: creatinine clearance (CrCl) 60 mL/min;
Satisfactory performance status: 2 by the Eastern Cooperative Oncology Group system.
Exclusion criteria included keratinizing squamous cell carcinoma or adenocarcinoma, age 70 years, pregnancy or lactation, history of previous treatment, or prior malignancy (except adequately treated carcinoma-in-situ of the cervix or basal/squamous cell carcinoma of the skin).
Pretreatment Evaluation and Randomization
All patients were evaluated by complete physical examination, fiberoptic nasopharyngoscopy, computed tomography or magnetic resonance imaging (MRI) of the nasopharyngeal region, chest x-ray, complete blood count, renal and liver function tests, and lactate hydrogenate (LDH). Additional investigations were performed only for those with suspicious findings or abnormal biochemical profile.
All patients were required to give written informed consent before registration. They were stratified by treatment center, T-category (T1-2 v T3–4), and N-category (N2 v N3). The blocked randomization scheme15 was used to allocate patients in equal proportions to the two treatment arms.
Radiotherapy
Patients in both arms were irradiated with megavoltage photons using the same RT technique and dose schedules in line with the treatment policy adopted by each center. All were treated with conventional fractionation of 2 Gy per fraction, 5 daily fractions per week, to a total dose 66 Gy.
Technique ranged from conventional two-dimensional technique to three-dimensional conformal or intensity-modulated technique throughout. All potential sites of local infiltration and bilateral cervical lymphatics were irradiated to 50 Gy. Additional boosts to the parapharyngeal space, the primary or nodal sites could be given (when indicated), the boost field was confined to the involved site with exclusion of critical structures, and the boost dose did not exceed 20 Gy.
Chemotherapy
Patients randomly assigned to the CRT arm were given additional chemotherapy using the IGS regimen. Cisplatin (CDDP) 100 mg/m2 was given intravenously every 3 weeks on days 1, 22, and 43 in concurrence with RT. Subsequent adjuvant chemotherapy using a combination of CDDP 80 mg/m2 intravenously and fluorouracil 1,000 mg/m2/d by 96-hour infusion was given every 4 weeks starting on days 71, 99, and 127.
Dose modifications were based on the nadir blood counts and interim toxicities of the preceding cycle. Chemotherapy was withheld until WBC 3,500/μL and platelet 100,000/μL. If the nadir of WBC was 2,500/μL and platelet 75,000/μL, full scheduled doses were given. If the nadir of WBC was 1,500/μL (less than 2,500/μL) or platelet 50,000/μL (less than 75,000/μL), CDDP was decreased by 20 mg/m2. If the nadir of WBC was less than 1,500/μL or platelet less than 50,000/μL, CDDP was decreased by 40 mg/m2 and fluorouracil by 200 mg/m2.
For renal toxicity, CDDP was withheld until CrCl 60 mL/min. If the nadir of CrCl was 40 (less than 60 mL/min), CDDP was decreased by 20 mg/m2 or replaced by carboplatin (using area under the curve of five). If the nadir of CrCl was less than 40 mL/min, or the patient developed otologic/neurologic toxicity grade 3, CDDP was stopped totally.
The dose of fluorouracil was decreased by 200 mg/m2 for grade 2, and 400 mg/m2 for grade 3 stomatitis or gastrointestinal toxicities, and stopped permanently if the patient developed drug-related angina.
Patient Assessments and Follow-Up
Acute RT-related toxicities were graded according to the Acute Radiation Morbidity Scoring Criteria of the Radiation Therapy Oncology Group16 and chemotherapy-related toxicities by the WHO criteria.17 Incidents of toxicities grade 3 (except nausea or alopecia) were recorded.
The first assessment of tumor response was performed 6 weeks to 16 weeks after completion of RT. All patients were assessed by complete physical examination and fiberoptic nasopharyngoscopy. Further investigations with computed tomography/MRI or other tests were arranged when indicated. Management of residual disease, if found, was decided by individual center. For statistical purposes, persistent primary/nodal disease at 16 weeks after completion of RT was taken as locoregional failure.
Patients were followed-up at least every 3 months during the first 3 years, and then every 6 months thereafter until death. The earliest dates of detecting tumor relapse at different sites were recorded. Salvage treatment was attempted, as far as possible, in line with the policies of individual centers. Late toxicities (occurring > 90 days after commencement of RT) were defined and graded according to the Late Morbidity Scoring Criteria of Radiation Therapy Oncology Group.16 The earliest date of detecting toxicity (except xerostomia and dental caries) grade 3 was recorded.
Statistical Methods
The primary end points of this study include failure-free survival (FFS), defined as the time to first failure at any site, and progression-free survival (PFS), defined as the time to first failure or death as a result of any cause. Assuming that the 5-year FFS by RT alone was 40%, the target accrual for detecting a 15% difference with 80% power was 340 patients.
Secondary end points for treatment efficacy included: locoregional failure-free rate (LR-FFR; persistence/recurrence in the nasopharyngeal and/or cervical region), distant failure-free rate (D-FFR; hematogenous metastasis), and overall survival (OS; death as a result of any cause).
Secondary end points for major treatment toxicities (grade 3) included crude incidence rates and the actuarial rate of late toxicities. For patients who had re-irradiation for treatment of locoregional relapses, events were censored at commencement of re-irradiation for assessing toxicities incurred solely by the primary treatment.
All events were measured from the date of registration, and the analyses were performed on an intention-to-treat basis. The actuarial rates were calculated by the product-limit method,18 and the differences compared by the Mantel-Haenszel log-rank test.19 The Breslow-Day 2 test was used for comparing incidence rates and categoric variables, and Student's t test was used for comparing the means of continuous variables. Multivariate analyses by Cox proportional hazards regression model20 were performed to test the independent significance of different factors. Covariates include host factors (ie, age, sex, performance status), tumor factors (ie, stage, LDH), RT (ie, total dose, overall treatment time, boost), and chemotherapy intervention (ie, treatment arm). All statistical tests were two-sided, and P values less than .05 were considered statistically significant. Our protocol specified two analyses to be performed, one at completion of accrual and another with complete 5-year follow-up.
RESULTS
Patient Characteristics
March 1999 to January 2004, 348 eligible patients were randomly assigned. Participating centers included Pamela Youde Nethersole Eastern Hospital, Tuen Mun Hospital, Queen Mary Hospital and Queen Elizabeth Hospital from Hong Kong, and Princess Margaret Hospital from Canada.
The two arms were well-balanced in all patient characteristics and tumor factors (Table 1). Ninety-five percent of patients had been regularly followed-up until death or the latest scheduled assessment. The median duration of observation was 2.3 years (range, 0.2 to 5.3).
Treatment and Compliance
The two arms were well-balanced in all RT parameters (Table 1). The techniques used were balanced (P = .87). The median total dose was 68 Gy, and overall time 46 days in both arms. Parapharyngeal boost via a postero-lateral field was given to 29% in the CRT arm and 28% in the RT arm (P = .89). Among those given additional boosts, the median dose given was 10 Gy. Only two patients in the CRT arm and one patient in the RT arm failed to complete the scheduled total dose; and only one patient in each arm had prolongation more than 7 days.
Four patients had major protocol violations: two patients in the CRT arm had no chemotherapy because of an incidental cause and the patient's refusal, and two patients in the RT arm were given chemotherapy as a result of rapid disease progression.
In the CRT arm, 65% of patients completed all six cycles of chemotherapy, and 79% had five cycles. The mean number of cycles given during the concurrent phase was 2.5 (± 0.6) and 2.6 (± 1.3) cycles during the adjuvant phase. Table 2 shows the details of compliance during each phase. The mean total dose of CDDP and fluorouracil given were 444 and 9,145 mg/m2 (82% and 76% of the total scheduled doses, respectively).
Acute Toxicities
Altogether 143 patients (84%) in the CRT arm had one or more toxicities, as compared to 93 patients (53%) in the RT arm (Table 3; P < .001). The majority of toxicities were uneventful and patients recovered rapidly, but the CRT arm had more grade 4 toxicities (12% v 1%) and one patient succumbed.
In addition to chemotherapy-related toxicities, the CRT arm also had significantly higher incidence of RT-related mucositis (62% v 48%, P = .01), but not of skin reaction (20% v 17%, P = .41).
The total incidence of chemotherapy-related toxicities in the CRT arm was 59%. The most common were leukopenia and/or neutropenia (32%); one patient died of neutropenic sepsis. The next most common chemotherapy-related toxicities included anemia (20%) and vomiting (19%). Besides the well-known toxicities, 2% had reactivation of hepatitis and 1% pulmonary tuberculosis.
Tumor Control
Only 2% in the CRT arm and 1% in the RT arm failed to achieve complete response at 16 weeks after completion of RT. At the time of analyses (September 2004), 39 patients (23%) in the CRT arm and 58 patients (33%) in the RT arm had failed at one or more sites (Table 4). The FFS showed significant improvement (Fig 1; 72% v 62% at 3-year, P = .027). The hazard ratio (HR) was 0.60; 95% CI, 0.40 to 0.92.
The CRT arm achieved significantly higher LR-FFR (Fig 2A; 92% v 82%, P = .005; HR = 0.28; 95% CI, 0.12 to 0.64). However, the improvement in D-FFR was insignificant (Fig 2B; 76% v 73%, P = .47; HR = 0.84; 95% CI, 0.53 to 1.34).
Among the patients with relapse, the proportion retreated aggressively was not significantly different: 59% (23 out of 39) in the CRT arm v 66% (38 out of 58) in the RT arm (P = 0.51). Salvage treatment given included chemotherapy (36% v 47%), re-irradiation (8% v 12%), and surgery (3% v 14%). The resultant salvage rate was higher in the RT arm, but the difference was insignificant (8% v 19%, P = .12).
Hence, the incidence of death as a result of disease progression (Table 4) was only slightly lower in the CRT arm (15% v 19%). However, there were extra deaths as a result of other causes (4.7% v 1.2%) in the CRT arm, including 1.1% treatment mortality, 1.1% unknown causes, and 3.5% incidental causes (ie, infection, cerebral vascular accident, suicide, and second malignancy), while only 1.1% in the RT arm died of incidental causes (ie, infection, dermatomyositis). The 3-year OS was 78% in both arms (Fig 3; P = .97; HR = 0.99; 95% CI, 0.62 to 1.58).
Altogether 46 patients (27%) in the CRT arm and 60 patients (34%) in the RT arm had experienced treatment failure or had died. Though the 3-year PFS was higher in the CRT arm (70% v 61%, P = .10), the trend of improvement was short of statistical significance (HR = 0.70; 95% CI, 0.47 to 1.04).
Subset analyses (Table 5) showed that the effect on D-FFR and OS was insignificant in all subsets. The greatest magnitude of differences was achieved among the 142 patients with stage IV disease (D-FFR, 62% v 52%; OS, 66% v 61%), but none of the end points reached statistical significance in this group.
On multivariate analyses (Table 6), CRT was an independent significant factor not only for LR-FFR and FFS, but also for PFS with hazard for progression being 0.59 (95% CI, 0.40 to 0.89; P = .012). However, the impact on OS remained insignificant with hazard for death being 0.79 (95% CI, 0.48 to 1.29; P = .34).
Among the other prognostic factors, total RT dose was significant for all end points including OS with hazard for death being 0.94 (95% CI, 0.92 to 0.97) per Gy increase. Both stage and LDH were significant for all end points except LR-FFR.
Late Toxicities
Thirty-four patients (20%) in the CRT arm and 20 patients (11%) in the RT arm developed one or more late toxicities attributed to the primary course of treatment (Table 7). The 3-year actuarial rate of late toxicities was higher in the CRT arm (Fig 4; 28% v 13%, P = .024; HR = 1.87; 95% CI, 1.08 to 3.26). This was mostly because of increased incidence of otologic toxicities (14% v 8%), peripheral neuropathy (2% v 0%), and endocrine dysfunction (4% v 1%). The majority of toxicities were grade 3 in severity. Only 3% in the CRT arm and 1% in the RT arm had grade 4 toxicities. Damages to neurologic structures were rare, but one patient (0.6%) in the CRT arm developed last 4 cranial nerves palsy and died of aspiration pneumonia.
DISCUSSION
Concurrent CRT is one of the most promising strategies for improving tumor control of advanced NPC, as this has potential for both enhancing the local effect of RT and eradicating micrometastases. There is little controversy that this is more potent than sequential chemotherapy, and is, thus far, the only strategy with significant survival benefit proven by randomized trials. However, with rapid advances in RT technologies, it is important that we provide patients with accurate information about the magnitude of benefit and the balance against excessive toxicities. The current trial, focused on patients with N2-3 disease, provides a good opportunity for studying the efficacy of the IGS regimen particularly for distant control. The findings are valid as the two arms were well-balanced in all prognostic factors and RT parameters (Table 1). Although the lack of routine imaging studies for exclusion of initial distant metastases and subsequent monitoring might lead to under-diagnosis of failures, this would not bias the current conclusions, as this concern applies to both treatment arms. No trends of potential bias caused by the bad news travels fast phenomenon21 are found in our data on additional testing. Furthermore, the sample size in the current study is large compared with most other trials.3,7,22,23 The only limitation is that the follow-up period (median, 2.3 years) is still relatively short, particularly for assessing late toxicities. Thus, the current conclusions should be taken as preliminary.
Table 8 summarizes our results together with other randomized trials on concurrent CRT3,7,22-24 to give a more comprehensive perspective of our current knowledge. It should be noted that there are variations in staging system, patients included in previous trials3,7,22,24 were actually staged II–IVB by the criteria of the current staging system,13,14 while the current trial only included the regionally-advanced group within stages III–IVB. Cross-series comparison is difficult.
The latest data published and/or presented at academic conferences are quoted to show the most updated data. In addition to the progress report by the Intergroup,4 a new trial presented by Wee et al23 also showed significant survival benefit by the IGS regimen (2-year OS, 85% v 77%, P = .02). However, it must be cautioned that the results currently listed are based on their abstracts published in the conference proceedings; full publications with thorough peer review are still pending.
The most recent publication is the progress report on the large series of 350 patients by Chan et al,25 comparing RT versus CRT using CDDP 40 mg/m2 weekly during the concurrent phase and no adjuvant therapy. With a median follow-up of 5.5 years, both FFS (the primary end point in their first report)24 and OS (the primary end point in their progress report)25 showed a positive trend short of statistical significance; the 5-year results for FFS were 60% v 52% (P = .16) and for OS were 70% v 59% (P = .065). Only in the Cox regression analysis, after adjusting for T-stage, overall stage, and age, did the difference in OS reach statistical significance (HR = 0.74; 95% CI, 0.5 to 1.0; P = .049).
Our data supports the conclusion by IGS3,4 and Lin et al7 that CRT could significantly improve locoregional control (Fig 2A; 92% v 82% at 3-years, P = .005). In concurrence with Lin et al, our primary end point FFS also achieved significant improvement (Fig 1; 72% v 62%, P = .027). The magnitude of difference in 3-year PFS (71% v 60%, P = .10) was substantially smaller than the original IGS report (69% v 24%),3 though multivariate analyses did show that CRT intervention was a significant factor for PFS (Table 6; HR = 0.59; 95% CI, 0.40 to 0.89; P = .012).
However, our data concur with the finding by Chan et al24,25 that distant control did not show significant improvement (Fig 2B; 76% v 73%, P = .47). Furthermore, with aggressive salvage after relapse, the OS were almost identical up to 3 years (Fig 3; 78% v 78%, P = .97). This phenomenon occurred in all subsets (Table 5). Though the 142 patients with stage IV disease did show greater magnitude of benefit in D-FFR (62% v 52%) and OS (66% v 61%), the differences did not reach statistical significance (P > .23).
Closer review of the trials with 5-year results for patients with stages II–IVB (by AJCC/UICC-5) showed that absolute survival gain achieved was 30% (67% v 37%, P < .01) by IGS,4 18% (72% v 54%, P = .002) by Lin et al,7 and 11% (70% v 59%, P = .065) by Chan et al.25 It is apparent that the exceptionally great magnitude of survival gain by IGS is mostly because of the poor result of the RT arm. One possible explanation is the difference in histologic pattern (28% of keratinizing type in the IGS as compared with < 4% in other series).
Our current failure to achieve significant survival benefit is probably not caused by suboptimal chemotherapy, because the dose of CDDP scheduled in our implementation of the IGS regimen was the highest among the different regimens used, and our compliance was not exceptionally poor (65% completed all six cycles, and the mean total dose of CDDP given was 444 mg/m2). Although CRT did cause significant increases in incidence and severity of toxicities both acute (Table 3; 84% v 53%, P < .001) and late (Table 5; Fig 4; 28% v 13% at 3-year, P = .024), the majority of toxicities were uneventful grade 3 toxicities. The IGS regimen could be accepted as reasonably well-tolerated even for Asian patients, but it should be noted that in addition to 1% treatment mortality (because of neutropenic sepsis and aspiration pneumonia), the CRT had an extra 4% mortality caused by unknown and incidental causes (including infection, cerebral vascular accident, and suicide).
Given the higher proportion of advanced cases in the current series (all being stages III–IVB by AJCC/UICC-5), it appears that the OS achieved by our CRT arm was not inferior to other series, but it was the favorable results of the RT arm that narrowed any potential gain. The obvious question is whether the results of the current RT arm are reproducible. To obtain the most representative data on treatment results in the modern era, the Hong Kong Nasopharyngeal Cancer Study Group has conducted a retrospective study of patients treated in all local public centers during 1996 to 2000.26 The technological provision was slightly inferior as only 32% of the series was staged by MRI and 10% irradiated by three-dimensional conformal or intensity-modulated technique (compared with 69% and 50%, respectively, in the current series). The 3-year OS for the 905 consecutive patients with stages III–IVB disease (by AJCC/UICC-5) treated by RT alone was 74%. Hence, the results achieved in the current trial should be reproducible, especially for nonkeratinizing/undifferentiated carcinoma. Indeed, even better results might be achievable, at least for locoregional control, with increasing availability of modern technologies.
In conclusion, our preliminary results confirm that CRT using the IGS regimen could improve tumor control, particularly at locoregional sites. It is possible that corresponding gain in survival may manifest with longer follow-up. We support recommending this option to patients with advanced NPC. However, patients should be duly informed that with improving RT technologies, the differential gain in survival might be small and there is risk of significant increase in toxicities. More potent therapy for distant control and more accurate prognostication to minimize unnecessary over-treatment are urgently needed. Estimation of sample size for future trials should be based on more realistic magnitude of benefit and higher baseline results by modern RT to avoid under-power.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank the data monitoring committee (K. Ang, P. Johnson, R. Chappell, and J. van Houwelingen), data manager (M.W. Chan), internal monitors (W.M. Ling, W. Sung, C.M. Lam, and J Yuen), et al of participating centers for their contribution.
NOTES
Supported by grants from the Hong Kong Cancer Fund, Ho Hung Chiu Medical Foundation Limited, and the Hong Kong Anti-Cancer Society, Hong Kong.
Presented in part at the 40th Annual Meeting of the American Society of Clinical Oncology, New Orleans, LA, June 5-8, 2004), and the 46th Annual Meeting of the American Society of Therapy Radiation Oncology, Atlanta, GA, October 3-7, 2004.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Parkin DM, Whelan SL, Ferlay J, et al: Cancer Incidence in Five Continents. IARC Scientific Publications Number 155, Vol. 8. Lyon, France, IARC Press, 2002, pp 534-536
Marks JE, Phillips JL, Menck HR: The National Cancer Data Base Report on the relationship of race and national origin to the histology of nasopharyngeal carcinoma. Cancer 83:582-588, 1998
Al-Sarraf M, LeBlanc M, Giri PG, et al: Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: Phase III randomized Intergroup study 0099. J Clin Oncol 16(4):1310-1317, 1998
Al-Sarraf M, LeBlanc M, Giri PGS, et al: Chemo-radiotherapy (CT-RT) vs radiotherapy (RT) in patients (PTS) with advanced nasopharyngeal cancer (NPC). Intergroup (0099) (SWOG 8892, RTOG 8817, ECOG 2388) Phase III Study: Progress report. J Clin Oncol 17:385a, 1998 (abstr 1483)
Beahrs O, Henson DE, Hutter RVP, et al: American Joint Committee on Cancer: Manual for Staging of Cancer, Fourth Edition. Philadelphia, PA, Lippincott, 1992, pp 33-38
Chow E, Payne D, O'Sullivan B, et al: Radiotherapy alone in patients with advanced nasopharyngeal cancer: Comparison with an intergroup study –Is combined modality treatment really necessary? Radiother Oncol 63:269-274, 2002
Lin JC, Jan JS, Hsu CY, et al: Concurrent chemoradiotherapy versus radiotherapy alone for advanced nasopharyngeal carcinoma: Positive effect on overall and progression-free survival. J Clin Oncol 21:631-637, 2003
Wolden SL, Zelefsky MJ, Kraus DH, et al: Accelerated concomitant boost radiotherapy and chemotherapy for advanced nasopharyngeal carcinoma. J Clin Oncol 19:1105-1110, 2001
Chua DTT, Sham J, Au G, et al: Concomitant chemoirradiation for Stage III-IV nasopharyngeal carcinoma in Chinese patients: Results of a matched cohort analysis. Int J Radiat Oncol Biol Phys 53:334-343, 2002
Le QT, Tate D, Koong A, et al: Improved local control with stereotactic radiosurgical boost in patients with nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 56:1046-1054, 2003
Lee N, Xia P, Quivey JM, et al: Intensity-modulated radiotherapy in the treatment of nasopharyngeal carcinoma: An update of the UCSF experience. Int J Radiat Oncol Biol Phys 53:12-22, 2002
Shanmugarantnam K, Sobin LH. Histological Typing of Tumors of the Upper Respiratory Tract and Ear, Second Edition. New York, NY, Springer-Verlag, 1991, pp 32-33
Fleming ID, Cooper JS, Henson DE, et al: American Joint Committee on Cancer: AJCC Cancer Staging Manual, Fifth Edition. Philadelphia, PA, Lippincott-Raven, 1997, pp 31-39
Sobin LH, Wittekind Ch. International Union Against Cancer (UICC): TNM Classification of Malignant Tumors, Fifth Edition. New York, NY, Wiley-Liss, 1997, pp 25-32
Freedman J, Furberg C, DeMets D. Fundamentals of Clinical Trials. New York, NY, Springer-Verlag, 1998
Cox JD, Stetz J, Pajak TF: Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 31:1341-1346, 1995
Miller AB, Hoogstraten B, Staquet M, et al: Reporting results of cancer treatment. Cancer 47:207-214, 1981
Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Peto R, Pike MC, Armitage P, et al: Design and analysis of randomized clinical trials requiring prolonged observation of each patient. Br J Cancer 35:1-39, 1977
Cox DR: Regression models and life tables. J Roy Stat Soc B 34:187-220, 1972
Schiffer CA, Dodge R, Larson RA: Long-term follow-up of Cancer and Leukemia Group B studies in acute myeloid leukemia. Cancer 80:2210-2214, 1997
Kwong DLW, Sham JST, Au GKH, et al: Concurrent and adjuvant chemotherapy for nasopharyngeal carcinoma: A factorial study. J Clin Oncol 22:2643-2653, 2004
Wee J, Tan EH, Tai BC, et al: Phase III randomized trial of radiotherapy versus concurrent chemoradiotherapy followed by adjuvant chemotherapy in patients with AJCC/UICC (1997) stage 3 and 4 nasopharyngeal cancer of the endemic variety. J Clin Oncol 23:487, 2004 (abstr 5500)
Chan ATC, Teo PML, Ngan RK, et al: Concurrent chemotherapy-radiotherapy compared with radiotherapy alone in locoregionally advanced nasopharyngeal carcinoma: Progression-free survival analysis of a phase III randomized trial. J Clin Oncol 20:2038-2044, 2002
Chan AT, Leung SF, Ngan RK, et al: Overall survival after concurrent cisplatin-radiotherapy compared with radiotherapy alone in locoregionally advanced nasopharyngeal carcinoma. J Natl Cancer Inst 97(7):536-539, 2005
Lee AW, Sze WM, Au JS, et al: Treatment results for nasopharyngeal carcinoma in the modern era: The Hong Kong experience. Int J Radiat Oncol Biol Phys 61(4):1107-1116, 2005(Anne W.M. Lee, W.H. Lau, )