- © 2005 by American Society of Clinical Oncology
Concomitant Boost Radiation Plus Concurrent Cisplatin for Advanced Head and Neck Carcinomas: Radiation Therapy Oncology Group Phase II Trial 99-14
- K. Kian Ang,
- Jonathan Harris,
- Adam S. Garden,
- Andy Trotti,
- Christopher U. Jones,
- Luis Carrascosa,
- Jonathan D. Cheng,
- Sharon S. Spencer,
- Arlene Forastiere and
- Randal S. Weber
- From The University of Texas M.D. Anderson Cancer Center, Houston, TX; Radiation Therapy Oncology Group Statistical Unit, University of South Florida H. Lee Moffitt Cancer Center, Tampa, FL; Radiological Associates of Sacramento, Sacramento, CA; University of Louisville, Louisville, KY; Fox Chase Cancer Center; University of Pennsylvania Medical Center, Philadelphia, PA; University of Alabama at Birmingham, Birmingham, AL; and The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Baltimore, MD
- Address reprint requests to K. Kian Ang, MD, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Box 97, Houston, TX 77030; e-mail: kianang{at}mail.mdanderson.org
Abstract
Purpose To investigate the feasibility of combining concomitant boost accelerated radiation regimen (AFX-C) with cisplatin and to assess its toxicity and the relapse pattern and survival in patients with advanced head and neck carcinoma (HNC).
Patients and Methods Between April and November of 2000, 84 patients with stage III to IV HNC who met the eligibility criteria were enrolled; 76 of these patients were analyzable. Radiation consisted of 72 Gy in 42 fractions over 6 weeks (daily for 3.5 weeks, then twice a day for 2.5 weeks). Cisplatin dose was 100 mg/m2 on days 1 and 22. Tumor and clinical status were assessed, and acute late toxicities were graded.
Results Sixty-five patients (86%) received both radiation and chemotherapy per protocol or with minor variations. The estimated 2-year locoregional relapse and distant metastasis rates were 34.7% and 16.1%, respectively. The estimated 2-year overall survival and disease-free survival rates were 71.6% and 53.5%, respectively. Three patients (4%) died of complications, 19 patients (25%) had acute grade 4 toxicity, and 49 patients (64%) had acute grade 3 toxicity. The 2-year cumulative incidence of late grade 3 to 5 toxicities was 51.3%.
Conclusion These data showed that it was feasible to combine AFX-C with cisplatin. The compliance to therapy was high, and the locoregional control and survival rates achieved compared favorably with AFX-C alone or other concurrent chemoradiation regimens tested by the Radiation Therapy Oncology Group. A phase III trial comparing AFX-C plus cisplatin against standard radiation plus cisplatin is ongoing to determine whether the use of AFX-C in the concurrent chemoradiation setting further improves outcome.
INTRODUCTION
There has been intensive clinical research aimed at improving locoregional tumor control and functional and cosmetic outcome for patients with advanced head and neck carcinomas (HNC) by rational modification of radiation fractionation regimen and by combining radiation with chemotherapy. The two groups of biologically sound modified fractionation regimens that have been extensively studied are hyperfractionation and accelerated fractionation. 1, 2 Hyperfractionation stemmed from the observation of preferential sparing of late responding tissues relative to epithelial tissues and some tumors as a results of decreasing the size of radiation dose per fraction, and accelerated fractionation regimens emerged through the recognition of the magnitude and hazard of tumor clonogen proliferation during the course of radiotherapy. 3, 4 Results of large randomized trials addressing the optimization of radiation fractionation regimen collectively show that a number of biologic-based modifications of fractionation schedules have improved the locoregional control rate in the order of 10% to 15% but had only a modest impact on the overall survival rate. 1 Although several altered fractionation regimens consistently induce more severe acute mucositis, the general observation is that the late toxicity is not appreciably increased.
Results of scores of clinical trials testing combined-modality therapy have been published. Meta-analysis of the data revealed that, in aggregate, cytotoxic agents administered before (induction or neoadjuvant chemotherapy) or after (adjuvant chemotherapy) surgery or radiation do not improve the therapeutic outcome appreciably over locoregional treatment alone. In contrast, chemotherapy administered concurrently with radiation has improved the 2- and 5-year overall survival rates by 8% 5 but, unfortunately, at the expense of increased toxicity. 6 The data of five recent trials addressing standard radiation fractionation with or without cisplatin have been reported. 7- 11 All five trials showed superior outcome in favor of the combined regimen in terms of locoregional control or organ preservation, and three trials also showed improvement in survival. 7, 8, 11 Six other trials also show a varying degree of benefits of combining various radiation schedules concurrently with other chemotherapy regimens. 12- 17 Despite these consistent data, the best regimen to be recommended for the treatment of patients with advanced HNC remains unclear.
A phase III trial of the Radiation Therapy Oncology Group (RTOG 90-03), which enrolled 1,113 patients, revealed that both the hyperfractionation regimen and accelerated fractionation by concomitant boost regimen (AFX-C) yielded significantly better locoregional control than standard fractionation in patients with advanced HNC. 18 Building on these findings, a phase II study was designed to test the combination of AFX-C with cisplatin in patients with advanced HNC. Single-agent cisplatin was chosen because it had been found to improve outcome when combined with conventionally fractionated radiotherapy in five phase III cooperative group trials. 7- 11
PATIENTS AND METHODS
Study Objectives and Patient Eligibility
Patients with previously untreated locally advanced (stage III or IV) squamous cell carcinoma of the oral cavity, oropharynx, hypopharynx, or larynx who had a Zubrod performance status of 0 to 1 and life expectancy of at least 6 months, met the blood count and liver function test criteria, and elected to receive nonsurgical primary therapy were enrolled. Patients younger than 18 years of age or with a prior (within 5 years) or synchronous malignancy other than nonmelanoma skin cancer were excluded. A medical history and physical examination, CBC, chest x-ray, computed tomography or magnetic resonance imaging of the head and neck, diagram of the primary tumor and neck nodes, and dental evaluation were required. The disease was staged according to the 1998 classification of the American Joint Committee on Cancer staging.
Treatment
Radiotherapy was delivered in 1.8 Gy per fraction, 5 fractions a week, to 54 Gy in 30 fractions over 6 weeks to the initial target volume encompassing gross tumor and clinically and radiologically involved nodes along with regions of potential contiguous and lymphatic spread. At 32.4 Gy in 18 fractions (ie, latter part of week 4), a second daily dose of 1.5 Gy per fraction (with at least a 6-hour interval) was administered to the boost volume covering gross tumor and involved nodes for a total of 18 Gy in 12 treatment days. The primary tumor and clinically and radiologically involved nodes received 72 Gy in 42 fractions over 6 weeks, and uninvolved nodes received 54 Gy in 6 weeks. Clinically and radiologically negative posterior and lower neck nodes received a minimum dose of 50.4 Gy at 3 cm. A combination of lateral-opposed portals, anterior and lateral wedged fields, or other field arrangements was used to treat the primary tumor and the lymph nodes in the upper neck. A single anterior field was used to treat the neck below the fields for the primary tumor. All fields were treated on each treatment day.
Cisplatin was administered in a dose of 100 mg/m2 intravenously on days 1 and 22 with granisetron or ondansetron premedication and vigorous hydration and diuresis. Guidelines for dose modification because of cytopenia, neurotoxicity, or nephrotoxicity were specified in the protocol. Neck dissection was allowed for patients with multiple neck nodes or with lymph nodes exceeding 3 cm in diameter (ie, N2 and N3) even when complete nodal response was achieved. Neck dissection was required for patients with a palpable or suspicious radiographic abnormality persisting 6 weeks beyond completion of radiation and chemotherapy. 19
Follow-Up and Data Analysis
Patients underwent weekly examination during treatment. First follow-up evaluation occurred around 4 weeks after completion of therapy. Subsequently, patients were assessed every 3 months for the first 2 years, every 6 months in years 3 to 5, and annually thereafter. In addition to tumor and clinical status, acute and late (occurring > 90 days from start of treatment) normal tissue effects were graded. Systemic and acute radiation effects were scored using the National Cancer Institute Common Toxicity Criteria version 2.0, whereas late radiation effects were scored according to the RTOG/European Organisation for Research and Treatment of Cancer criteria. 20
The primary end point of the study was the locoregional failure rate at 1 year. Additional end points included toxicity rates, overall and disease-free survival rates, and the incidence of distant metastases. All time-to-failure end points were calculated from the date of registration to the study. Survival and disease-free survival rates were estimated using the Kaplan-Meier method, 21 whereas the rates of locoregional failure and distant metastases were calculated using the method of cumulative incidence 22 because this method accounts for competing risk (ie, death without disease relapse). Cox proportional hazards models 23 were used to compare survival from this regimen to historical controls.
RESULTS
Study Population and Compliance to Therapy
Between April and November of 2000, 84 patients were registered. Eight patients were excluded because they did not meet the eligibility criteria (four patients had initial surgical treatment, one had laryngeal carcinoma before it became eligible, and one had a platelet count of < 100,000 cells/mm3), did not receive any protocol therapy (one patient), or had delinquent data (one patient). Tables 1 and 2 list pretreatment patient and tumor characteristics. Of the 76 analyzable patients, 67 (88%) had stage IV disease.
Distribution of Patient and Tumor Characteristics
Distribution of Tumor and Node Stage
The overall radiation delivery was scored by the study chair as per protocol in 55 patients (72%) or with minor variation in 16 patients (21%). Only one patient was scored as having an unacceptable deviation from protocol because of a marginal miss, whereas four patients had incomplete radiotherapy because of patient refusal (two patients, 45.6 Gy and 62.4 Gy), tumor progression (one patient, 42.3 Gy), or protocol treatment–related death (one patient, 67.2 Gy). The fractionation regimen was according to protocol specification in 72 patients (95%). The duration of radiotherapy was ≤ 46 days in 59 patients (78%) and 47 to 51 days in 10 patients (13%).
Seventy patients (92%) received both cycles of cisplatin. Second cisplatin cycle was withheld in four patients or the dose was reduced by 20% or more in five patients because of toxicity. One patient refused the second cycle, and chemotherapy data was missing for another patient. Overall, 65 patients (86%) received therapy per protocol or with minor variation for both radiation and chemotherapy.
Tumor Response and Survival
Complete response to therapy was recorded in 63 patients (83%). Relapse at the primary site, at the regional nodes, and at both sites has occurred in eight, three, and three patients, respectively. Of the 13 patients with persistent tumor or residual changes, five had primary tumor progression, one had nodal progression, and two had disease progression at both sites. No tumor progression was detected in subsequent follow-up visits in the remaining five patients.
Twenty-nine patients had persistent nodal disease at their first posttreatment follow-up visit and should have undergone neck dissection. Of these 29 patients, 16 actually underwent neck dissection 6 to 32 weeks after completion of radiation and chemotherapy. Results of histologic examination were available in 15 patients (ie, no residual tumor in 10 patients, residual tumor with clear margin in three patients, and involved margin in two patients). In addition, eight patients underwent neck dissection that was not mandated per protocol 6 to 73 weeks after completion of radiation and chemotherapy. Five patients were N2-3 at presentation (no residual tumor in all five patients), two were N1 (no residual tumor in one patient), and one was N0 (no tumor). Of the total 24 patients who underwent neck dissection, only three have relapsed. A total of 14 patients (18%) developed distant metastasis, most frequently to the lung (11 patients).
At the time of analysis, 51 patients were alive, of whom 38 had no evidence of disease. The median follow-up for surviving patients is 2.2 years (range, 0.9 to 2.7 years). Twenty-five (33%) of 76 patients have died. The causes of death were tumor progression (n = 18), treatment toxicity (n = 3), intercurrent disease (n = 2), and uncertain (n = 2). No second primary tumors have been reported.
Figure 1 shows the actuarial overall and disease-free survival data along with cumulative incidence of locoregional failure and distant metastasis. Overall, 13 patients (17%) had persistent locoregional disease, eight (11%) experienced first relapse at the primary site, three (4%) experienced first relapse at the regional site, two (3%) experienced first relapse at both primary and nodal sites, and nine (12%) experienced first relapse at distant organs. First progression occurred after 2 years in only two patients (one nodal recurrence and one distant spread). The estimated 1- and 2-year rates of locoregional relapse, overall survival, disease-free survival, and distant metastasis were 29.2% (95% CI, 18.8% to 39.6%) and 34.7% (95% CI, 23.8% to 45.7%), 81.5% (95% CI, 72.7% to 90.3%) and 71.6% (95% CI, 61.3% to 81.9%), 60.4% (95% CI, 49.3% to 71.4%) and 53.5% (95% CI, 42.2% to 64.8%), and 11.9% (95% CI, 4.5% to 19.2%) and 16.1% (95% CI, 7.7% to 24.5%), respectively.
Overall and disease-free survival and relapse pattern of patients with advanced head and neck cancer treated with accelerated fractionation by concomitant boost regimen plus concurrent cisplatin. Forward slashes (/) indicate patients censored alive.
Acute and Late Toxicity
Three patients (4%) died of treatment-induced sepsis, pneumonia with acute respiratory distress syndrome, and/or renal failure. Two of the lethal events occurred during the acute period. In addition, 19 patients (25%) experienced acute grade 4 side effects, and 49 patients (64%) had acute grade 3 toxicity. The most common acute grade 4 to 5 side effects in descending frequencies, occurring in two or more patients, were leukopenia and infection (n = 7), dysphagia (n = 4), electrolyte imbalance (n = 4), anorexia and vomiting (n = 3), mucositis (n = 2), creatinine elevation (n = 2), and respiratory distress syndrome (n = 2). Grade 3 and 4 mucositis occurred in 38 patients (50%) and two patients (3%), respectively. Twenty-five patients developed cisplatin-related dehydration, which required brief intravenous fluid replenishment in 12 patients and several days of rehydration in 13 patients. Table 3 lists the type and frequency of acute side effects.
Type and Frequency of Acute Side Effects Observed in 76 Patients
Table 4 lists the type and frequency of severe (grade 3 to 5) late side effects. Only one patient died 102 days from the start (or 53 days from completion) of therapy as a result of renal failure and infection. Grade 3 and 4 late (> 90 days) side effects were recorded in 26 patients (36%) and 11 patients (15%), respectively. The most common type of late grade 4 late complication was mucosal ulceration (n = 4). Figure 2 shows the respective cumulative incidence curves of late grade 3 and grade 4 to 5 toxicity in 73 patients who survived more than 90 days. The 2-year cumulative incidences of late grade 3 to 5 and grade 4 to 5 toxicities were 51.3% (95% CI, 39.6% to 63.0%) and 16.6% (95% CI, 7.9% to 25.2%), respectively.
Cumulative incidence of all treatment-related late grade 3 to 5 and grade 4 to 5 toxicity. Forward slashes (/) indicate patients censored alive.
Type and Frequency of Late Side Effects Observed in 76 Patients
Of the 56 patients who did not have gastrostomy before treatment, 44 (79%) received gastrostomy for supplemental feeding. As of the last follow-up visit, 31 of these patients (55%) were alive with no evidence of disease, and nine of them still had a gastrostomy tube. Of the 20 patients who had gastrostomy before therapy, which reflects the advanced T stage, seven (35%) were alive with no evidence of disease, and only two were gastrostomy dependent for feeding. In summary, of the 38 patients alive without evidence of disease as of the last follow-up visit, 11 (29%) still had a gastrostomy tube.
DISCUSSION
The findings that a number of modified radiation fractionations and concurrent chemoradiation regimens are more effective than conventionally fractionated radiotherapy in the treatment of advanced HNC generated interest to test the combinations of altered fractionation regimens with chemotherapy. In a retrospective study, for example, Wolden et al 24 compared the data of 50 patients with nasopharyngeal carcinoma who had received concomitant boost radiation with two cycles of concurrent cisplatin (plus cisplatin-based adjuvant chemotherapy in most cases) with the data of an earlier cohort of 51 patients matched for prognostic factors who were treated with radiotherapy alone. They showed that the locoregional control, progression-free survival, and overall survival rates were better in the combined treatment group. Of note is that the regimen used was comparable to the regimen tested in this RTOG trial.
The results of six phase III trials testing the efficacy of such combinations against radiation alone have been reported. The radiation regimens used were accelerated fractionations in three trials, 17, 25 hyperfractionation in one study, 16 and split-course altered fractionation in two trials. 12, 13, 26 The results of these trials are listed in Table 5. Collectively, most trials show that combinations of modified fractionation regimens with chemotherapy achieve better local control and, in several trials, improved survival compared with standard or altered fractionation alone. However, the value of altered fractionation in the concurrent chemoradiation setting (ie, the potential benefit of combining altered fractionations instead of standard fractionation with chemotherapy) has not been tested.
Phase III Trials Addressing AFCT in Patients With Head and Neck Cancer
Building on the results of RTOG 90-03, which show locoregional tumor control benefit by concomitant boost regimen (AFX-C), 18 a phase II trial was undertaken to determine the feasibility of delivering two cycles of cisplatin (100 mg/m2) on days 1 and 22 of AFX-C in a cooperative group setting. The data revealed a high compliance rate to the protocol treatment, with 86% of patients completing both the radiation and chemotherapy as specified or with minor variations. The acute toxicity of the treatment was rather severe as expected. However, the toxic death rate was within the range of 2% to 5% observed in other intergroup trials testing sequential or concurrent cisplatin-based regimens combined with radiation. 8, 9 Comparing the acute and particularly late toxicity with trials conducted by other cooperative groups, however, is rather difficult because of inconsistency in recording and reporting, as clearly pointed out by Trotti and Bentzen. 27 These authors noted that four different recognized grading systems and two descriptive efforts had been used in reporting the results of nine frequently cited trials addressing the combination of radiation and chemotherapy in HNC published within the last decade. Comparison of the results of the current study with those of two preceding trials coordinated by the RTOG testing radiation with concurrent cisplatin (RTOG 91-11) 9 or three types of two-agent regimens (ie, cisplatin-fluorouracil, paclitaxel-cisplatin, or hydroxyurea-fluorouracil; RTOG 97-03) 28 revealed similar incidences of grade 3 to 4 adverse effects.
In term of reporting late toxicity, Trotti and Bentzen 27 found that only four of nine trials used a recognized grading system, two provided only descriptive information, three did not state late effects at all, and only one reported swallowing function and feeding tube dependence. When provided, only crude prevalence rates were reported rather than the actuarial rates or cumulative incidences, which are more representative because these figures account for patient loss to competing risk. The cumulative incidences of late grade 3 to 5 and 4 to 5 complications observed in the current series were 51.3% and 16.6%, respectively. Whether these rates are higher than those observed with other combined regimens is difficult to ascertain because the reported crude estimates of late effects range from 14% to 82%. Only the ongoing RTOG phase III trial (see final paragraph) can resolve whether the combination of altered fractionation with cisplatin induces more late effects than standard fractionation plus cisplatin.
The chronic swallowing dysfunction and feeding tube dependence observed in this study most likely result from the combination of normal tissue destruction secondary to tumor infiltration and treatment effect because 20 patients (26%) in this series already had gastrostomy before treatment. Of note is that five of the long-term survivors in this subgroup were able to regain swallowing function after achieving locoregional tumor control. The overall feeding tube dependence rate of 29% is at the lower range of the rates observed by Staar et al. 17
In light of 88% of the patients having stage IV disease, the 2-year overall survival rate of 71.6% and 2-year locoregional failure rate of 34.7% are quite encouraging when compared with a historical database consisting of similar patients from RTOG 90-03 and 97-03 using RTOG recursive partitioning analysis class 29 to adjust for prognosis. Therefore, it was decided to proceed with a phase III trial (RTOG H0129) to compare the efficacy and toxicity of AFX-C plus cisplatin with the efficacy and toxicity of standard fractionation plus cisplatin to determine whether AFX-C can yield an additional therapeutic benefit in the concurrent chemoradiation setting.
Appendix
The following Radiation Therapy Oncology Group institutions enrolled the majority of patients: University of Texas M.D. Anderson Cancer Center, University of South Florida H. Lee Moffitt Cancer Center, Radiological Associates of Sacramento, University of Louisville, Fox Chase Cancer Center, Virginia Mason Medical Center, and Cancer Research for the Ozarks.
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Consultant/Advisory Role: K. Kian Ang, Bristol-Myers Squibb. Research Funding: Randal S. Weber, RTOG. For a detailed description of these categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
Acknowledgments
We thank all clinical investigators for enrolling patients onto this trial. In addition, we acknowledge the dedication and hard work of the statisticians, clinical research associates, dosimetrists, and administrative staff who have contributed to the success of this trial. In particular, we are grateful for the valuable contributions of Thomas F. Pajak, PhD, in statistics, Rebecca Allegretto in data management, Bernadine Dunning in radiation quality assurance, and Cathy Ramirez in manuscript preparation.
Footnotes
-
Supported by grant Nos. U10 CA21661, U10 CA37422, U10 CA32115, and CA06294 awarded by the National Cancer Institute (Bethesda, MD) supplemented by the Gilbert H. Fletcher Chair.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
- Received December 8, 2003.
- Accepted October 25, 2004.
