Radiation therapy is the only curative treatment modality for nonmetastatic squamous cell carcinoma of the nasopharynx. Concurrent chemoradiation is the standard treatment strategy for nasopharyngeal cancer (NPC) in locally advanced stages (defined as a tumor classification of T3/T4 or positive lymph node [N+] status).[1-6] However, the results after such treatment are suboptimal. Clearly, novel treatment strategies are needed to further improve patients' survival rates.
The addition of a taxane to cisplatin and 5-fluorouracil (5-FU) (ie, the TPF regimen) before radiotherapy or concurrent chemoradiation has resulted in significantly improved treatment outcomes, including overall survival, in patients with head and neck squamous cell carcinoma (HNSCC), excluding NPC.[7-9] Such synergistic effects provided by TPF chemotherapy are equally attractive for its use in locoregionally advanced NPC. Therefore, we initiated 2 prospective phase 2 clinical trials in 2007—1 for stage III NPC and another for nonmetastatic stage IV NPC—to evaluate the efficacy and safety of induction TPF chemotherapy followed by concurrent chemoradiation using 3-dimensional (3D)-conformal radiotherapy (3D-CRT) or intensity-modulated radiotherapy (IMRT). Here, we report the 3-year outcomes from those trials.
MATERIALS AND METHODS
Patients who had histologically proven, nonkeratinizing, poorly differentiated and undifferentiated squamous cell carcinoma of the nasopharynx in stage III and IVA/IVB according to the 2002 American Joint Cancer Committee (AJCC) staging criteria were eligible for the 2 prospective phase 2 clinical trials. Both trials had identical treatment regimens, were approved individually by the institutional review boards, and were registered online (National Clinical Trial [NCT] 00816855 for stage III NPC and NCT 00816816 for stage IVA/IVB NPC). Written informed consent was required for all participating patients.
In addition to disease stage at diagnosis, other eligibility criteria included: age >18 years and <70 year; an expected lifespan of at least 6 months; a Karnofsky performance status score ≥70; and adequate bone marrow reserve (neutrophil count >2000/L, platelet count ≥100,000/L, and hemoglobin ≥10 g/dL), hepatic function (total bilirubin <1.5 mg/dL, aspartate and alanine aminotransferase levels <2 times the upper limits of normal), and renal function (creatinine clearance rate >50 mL per minute).
Exclusion criteria included: pathology types other than World Health Organization grade 2/3 NPC; stage IVC (ie, metastatic) disease; previous radiotherapy or surgery to the head and neck for any malignant or nonmalignant disease(s); a history of malignant tumors or of synchronous or multiple tumors, except for basal cell carcinoma of the skin; a positive pregnancy test for women; accompanying disease or status influencing the normal enrollment of patients or safety during the research; active mental disorders or other mental disorders that influenced the patient's signature on the informed consent or comprehensibility; uncontrolled, active infections; and participation in other clinical research at the same time.
A biopsy from the primary site was required in all patients for pathologic diagnosis. Pretreatment staging evaluations included clinical examination of the head and neck, magnetic resonance imaging (MRI) scans of the head and neck, a computed tomography scan of the thorax, a whole-body bone scan, abdominal sonography, a complete blood count, and a serum biochemical profile.
Patients received radiotherapy using 3D-CRT or IMRT techniques. The gross tumor volume (GTV) was defined as all known gross disease determined from clinical and imaging examinations, including the primary tumor (GTV-P) and metastatic lymph nodes (GTV-LN). The clinical target volume (CTV) for high-risk (CTV-H), subclinical disease consisted of a 5-mm margin surrounding the GTV, the whole nasopharyngeal cavity, the anterior 1 third or 2 thirds of the clivus (when invaded, the whole clivus was covered), the skull base, the pterygoid plates, the parapharyngeal space, the inferior sphenoid sinus (the whole sphenoid sinus had to be covered for T3 and T4 disease), the posterior of the nasal cavity and the maxillary sinus, and areas of lymph node drainage of the upper neck (the retropharyngeal lymph nodes and levels II, III, and Va). The low-risk CTV (CTV-L) referred to levels IV and Vb without metastatic cervical lymph nodes. Level Ib had to be included in the CTV-H for patients who had metastatic lymph nodes at level IIa and any lymph node drainage pathways that contained metastatic lymph nodes. A minimum of 5 mm around the GTV-P and 3 mm around the GTV-LN and CTVs were required in all directions to define each respective planning (P-) target volume (P-GTV-P, P-GTV-N, P-CTV-H, and P-CTV-L).
In 3D-CRT, P-GTVs and P-CTV-H received radiation doses of 70 grays (Gy) in 35 fractions and 60 Gy in 30 fractions, respectively. Because of the shortage of machine resources, for most patients who received IMRT, a simplified IMRT technique was used to shorten the radiation time in each fraction, with prescription doses of 70 Gy, 68.25 or 66.50 Gy, and 61.25 Gy (in 35 fractions) for the P-GTV-P, P-GTV-N, and P-CTV-H, respectively. The low neck or supraclavicular field (P-CTV-L) was treated with anteroposterior/posteroanterior fields and received 30 fractions at 1.8 Gy per fraction for a total of 54 Gy. Radiotherapy was delivered once daily in 5 fractions per week over 7 weeks. Some patients received boost irradiation doses of 4 to 6 Gy in 2 or 3 fractions to the residual focus in the nasopharynx or the neck.
Neoadjuvant chemotherapy with the TPF regimen before radiotherapy included docetaxel 75 mg/m2 on day 1, cisplatin 75 mg/m2 on day 1, and 5-FU 2500 mg/m2 as an intravenous infusion over 120 hours every 3 weeks for 3 cycles. Dose modifications were based on the preceding cycle nadir blood counts and interim toxic effects. A reduction in the docetaxel dose by 20% with constant cisplatin and 5-FU doses was allowed if a grade 4 hematologic adverse event or febrile neutropenia emerged in the previous course, and 5-FU was stopped if grade >3 mucositis or diarrhea occurred in the former course. Recombinant granulocyte-colony–stimulating factor (G-CSF) support was permitted for prophylaxis, and antibiotic therapy was used when patients had grade 4 neutropenia.
Concurrent chemoradiation consisted of weekly cisplatin (40 mg/m2) during radiotherapy for a maximum of 7 cycles, beginning on the first day of radiotherapy as planned. Chemotherapy at the full dose would be delivered strictly according to the treatment protocol, and no adjustment would be allowed under any situation. The start of chemotherapy could be postponed if the neutrophil count dropped to <2000/μL or the platelet count was <100,000/μL. Chemotherapy could be suspended if the creatinine clearance rate became <50 mL per minute.
After the completion of concurrent chemoradiation, all patients were assessed every 3 months during the first 3 years, every 6 months for the next 2 years, and annually thereafter. All local recurrences were diagnosed with nasopharyngoscopy and biopsy and an MRI of the head and neck. Regional recurrences were diagnosed by clinical examination of the neck and, in doubtful cases, with fine-needle aspiration or an MRI scan of the neck. Distant metastases were diagnosed by clinical symptoms, physical examinations, and imaging methods. Whenever necessary and possible, salvage treatments, including reirradiation, chemotherapy, and surgery, were provided to patients who developed recurrent disease.
The main objective of both trials was to evaluate the efficacy and safety of neoadjuvant chemotherapy with the TPF regimen followed by concurrent chemoradiation for stage III and stage IVA/IVB NPC. The primary endpoints for both trials were overall survival (OS); and the secondary endpoints were disease progression-free survival (PFS), local progression-free survival (LPFS), regional progression-free survival (RPFS), and distant metastasis-free survival (DMFS).
The OS rates at 5 years for patients with stage III and stage IVA/IVB NPC who received concurrent chemoradiation were approximately 65% and 45%, respectively. We projected a 20% OS improvement from 65% to 85% for patients with stage III disease and from 45% to 65% for patients with stage IVA/IVB disease after neoadjuvant chemotherapy with the TPF regimen followed by concurrent chemoradiation. Forty-seven patients were required for a Type I error rate of .05 (1-sided) with 80% statistical power to detect an increase of 20% in 5-year OS for the stage III trial, and 58 patients were required to achieve the same statistical power for the stage IVA/IVB trial. After adjusting for a 10% rate of dropout or loss to follow-up, the stage III NPC trial required 52 patients, and the stage IVA/IVB NPC trial required 64 patients.
Descriptive statistics were used to summarize the patient characteristics. The estimated OS, PFS, LPFS, RPFS, and DMFS rates were calculated using the Kaplan-Meier method, and differences between survival curves were assessed using the log-rank test in both studies. Survival duration was measured from the time of recruitment until either death or the date of the last follow-up visit for patients who remained alive.
The primary objective of the 2 phase 2 trials we are reporting here addressed the efficacy of induction docetaxel, cisplatin, and 5-FU (the TPF regimen) used with a standard chemoradiation regimen for stage III and stage IVA/IVB NPC. In particular, we focused on OS as the primary endpoint for patients who received the triple combination of TPF followed by concurrent chemoradiation. The results demonstrated a significant improvement in OS to 94.8% for patients with stage III NPC and to 90.2% for patients with nonmetastatic stage IV NPC after 3 years of follow-up. Furthermore, the triple combination of TPF was well tolerated when used before the standard chemoradiation strategy for NPC.
NPC is sensitive to both radiation and chemotherapy, and combined chemoradiation is the mainstay of treatment for advanced, nonmetastatic disease. The use of chemotherapy with radiation has been studied extensively. The results from several randomized phase 3 trials and meta-analyses have confirmed that adjuvant chemotherapy provides no significant improvement in patients' survival.[13-16] Induction cisplatin chemotherapy, when used before radiation alone, does improve local control and disease-free survival in patients with locally advanced NPC without affecting OS.[17-20] The receipt of radiation therapy and concurrent cisplatin-based chemotherapy with or without adjuvant chemotherapy, became the standard treatment of choice for T3 of T4 and/or N+ NPC after the publication of the pivotal Intergroup INT-0099 study and several confirming randomized trials.[2-6] However, adjuvant chemotherapy is poorly tolerated and has limited compliance. because patients suffer substantial toxicities from concurrent chemoradiation, and many are unfit and/or reluctant to receive further chemotherapy. Typically, only 60% of patients (range, 55%-76%) receive all 3 scheduled cycles of adjuvant chemotherapy, and 70% (range, 60%-81%) receive at least 2 cycles. The reported studies clearly illustrate that, with neoadjuvant chemotherapy, tolerance and compliance indeed are substantially better, and nearly 100% of patients (range, 97%-100%) can tolerate at least 2 cycles.
However, it has been suggested that the improvement in OS from concurrent chemotherapy is mostly because of the improvement in local disease control. On the basis of the published literature, the 5-year OS rates for patients with stage III and nonmetastatic stage IV NPC are approximately 65% and 45%, respectively. To further improve the outcome of patients with locally advanced NPC, an improvement in both local and distant control will be needed. However, induction chemotherapy using cisplatin or its combination, such as combined cisplatin, epirubicin, and paclitaxel, or adjuvant chemotherapy using cisplatin plus 5-FU failed to improve OS further according to the results from several randomized clinical trials.[13, 20, 22]
Docetaxel has demonstrated significant efficacy as a single agent or in combination with platinum in HNSCC. Compared with paclitaxel, it has less neurotoxicity, which supports its combined use with cisplatin. Its efficacy in the treatment of HNSCC, excluding NPC, has been demonstrated repeatedly in randomized clinical trials. The results from the TAX 323 and TAX 324 studies revealed that, when used with concurrent chemoradiation or radiation alone, the addition of docetaxel (T) to cisplatin and 5-FU (PF) (ie, TPF vs PF) reduced the risk of death by nearly 30%.
The effect of docetaxel used with cisplatin and without 5-FU (the TP regimen) for NPC has been addressed in 2 previously reported studies.[23, 24] A randomized phase 2 trial published in 2009 studied the toxicities and the primary outcome of patients with NPC who did or did not receive induction chemotherapy using the TP regimen followed by concurrent cisplatin chemotherapy and conventional radiation. In addition to the acceptable adverse-effect profile, the researchers demonstrated a significant improvement in OS for patients who received induction TP chemotherapy. The 3-year OS rate was 94.1% after induction chemotherapy followed by chemoradiation, compared with 65% after chemoradiation alone. This superb and exciting result was consistent with that reported in a more recently published, retrospective study: Ekenel et al reported their observation of 59 patients with NPC who received 3 cycles of induction cisplatin (75 mg/m2) and docetaxel (75 mg/m2) followed by combined cisplatin-based chemotherapy and conventional radiotherapy using cobolt-60 or a 4 to 6 megavolt linear accelerator. Impressive OS and PFS rates of 94.9% and 84.7%, respectively, were reported. In both studies, however, the numbers of patients in either stage were limited, especially those with stage IVA/IVB disease, who are at substantial risk for local and distant recurrence.
The findings from our trials confirm the encouraging results mentioned above and suggest that superior results are achieved in patients who receive TPF-based induction chemotherapy followed by concurrent chemotherapy and 3D-CRT or IMRT; ie, the current standard radiation technique. The local control rate in both studies reached >93%, including patients who were treated for T4 tumors. More important, the outcomes—especially OS and DMFS—after such treatment in patients with T4 and/or N3 disease are nearly identical to those observed in patients with stage III NPC. The OS and DFMS rates of approximately 90% in both stages indicate that TPF may substantially improve the control of systemic disease in patients with advanced NPC. The already initiated Radiotherapy Oncology Group for Head and Neck (GORTEC) multicenter phase 3 trial, in which patients are randomized to receive cisplatin plus radiation with or without neoadjuvant TPF chemotherapy (ie, TPF→concurrent cisplatin + RT vs cisplatin + RT), will confirm the efficacy of neoadjuvant TPF in a randomized fashion (registered online: http://www.clinicaltrials.gov/NCT00828386). However, that trial does not require IMRT, which is considered the current standard radiation technique for the treatment of NPC. Another ongoing randomized trial in collaboration with our institution will address the efficacy of neoadjuvant TPF (ie, TPF → concurrent cisplatin + IMRT vs cisplatin + IMRT) in patients with NPC who receive IMRT (registered online: http://www.clinicaltrials.gov/NCT01245959).
It has been reported that TPF induces a high rate of severe neutropenia (range, 76%-97%). In our studies, neutropenia occurred in 55.2% patients, which is less than the incidence of neutropenia reportedly caused by G-CSF support. However, the compliance with TPF neoadjuvant chemotherapy was better, all patients tolerated at least 2 cycles, and 88.8% of patients completed 3 cycles.
In a recently published, large, retrospective series of 370 patients, Lin et al demonstrated that OS after induction chemotherapy with TP or PF followed by IMRT without concurrent chemotherapy was as good as concurrent chemotherapy. If the hypothesis that neoadjuvant TPF significantly improves distant and local control is confirmed by the 2 ongoing randomized trials described above,[25, 26] then, based on the finding that IMRT improves the local control rate to >95% for all tumor classifications, whether concurrent chemotherapy (ie, the current standard) remains necessary becomes a question. This question (ie, TPF→concurrent cisplatin + IMRT vs TPF→IMRT) will be addressed further by a multi-institutional, randomized, noninferiority study that is in the planning stage; however, the initiation of that trial will depend on the results from the 2 ongoing randomized trials described above.[25, 26]
The results of our phase 2 trials have demonstrated that neoadjuvant TPF chemotherapy produces encouraging and potentially efficacious outcomes in terms of OS, PFS, and local control in patients with locoregionally advanced NPC. Furthermore, this regimen was well tolerated and had a manageable toxicity profile. However, this aggressive treatment combination remains investigational and can only be considered standard after confirmation by the already initiated phase 3 randomized trial.