Reviewers: John Han-Chih Chang, MD and Kenneth Blank, MD
Source: Journal of Clinical Oncology 1998; Volume 16 (Number 4): pages 1318 ? 1324.
Head and Neck cancer is a heterogeneous group of malignancies. This article addressed a wide range of sites within the head and neck ? from the oral cavity to the larynx and hypopharynx. Locally advanced disease (stage III and IV) can often times be difficult to resect due to invasion of critical vascular or neurological structures. Resection can also be so highly debilitating or deforming that it is not desirable to operate. For such cases, definitive radiotherapy (XRT) has been the modality of choice. It success in such situations has been varied from moderate to poor. Many have utilized accelerated hyperfractionation (AH) XRT (increased number of radiation treatments in a shorter total treatment time or duration). This has improved some of the results in some the sites in certain situations. Essentially, it remains controversial what patients truly benefit from this schema of treatment. Some have attempted to add chemotherapy (CTX) to improve outcome. This is the focus of this article from Munich, Germany.
Patients with squamous cell carcinoma of the oral cavity, floor of mouth, oropharynx, hypopharynx and larynx were enrolled. Patients were eligible if they were considered stage III or IV (T3 or greater and/or N1 or greater) and unresectable per the opinion of "an experienced head and neck surgeon." They further elucidated that this usually meant: cases in which surgery would not be possible without "mutilation." Resectable tumors that were not medical candidates for surgery were not eligible. Table 1 on page 1319 details the patient characteristics, which reveal that a great majority of the cases were of the oropharynx, hypopharynx and larynx. T4 (invasion into adjacent structures) and N2c (bilateral or contralateral lymph node metastases no greater than 6 cms) were the predominant stages of the diseases.
The patients were randomized to 2 arms: A) XRT alone versus B) CTX concurrent with XRT. The XRT was AH with scheduled breaks of one week every 2 weeks. Patients were treated twice a day with a 6-hour inter-fraction interval. One hundred and eighty centi-gray fractions were utilized. Patients started treatments on day 3 of week 1. Twice-a-day treatments continued to day 5. Seven fractions were given in week 2 (days 9 ? 12) in a twice-a-day fashion. Then a break was given in weeks 3 and 6 and the cycle of 13 fractions was given similarly in weeks 4 and 5 along as in weeks 7 and 8. The CTX agents used were cisplatin (CDDP), fluorouracil (5-FU) and leucovorin (LV). CDDP was 60 mg/m2 IV on day 2. 5-FU was 350 mg/m2 IV bolus day 2 and followed by 350 mg/m2/24 hours continuous infusion from day 2 ? 5. LV was 50mg/m2 IV bolus day 2 and followed by 100 mg/m2/24 hours continuous infusion from day 2 ? 5. This was repeated on every 21 days concurrent with the XRT.
Evaluation of response was at the 6-week interval from completion of treatment. No criteria were stated as far as what was considered a partial response versus complete. Neck dissection was only performed if the primary disease was considered a complete response.
From November 1989 to October 1993, 298 patients were enrolled of an anticipated 344 patients to achieve their of a significant survival advantage. The incomplete accrual was not mentioned or explained further. Twenty-eight patients withdrew consent after randomization, had a contraindication to CTX or had incomplete documentation.
In arm A, 80% of the 140 patients completed the XRT as scheduled. Acute toxicity in the remaining patients prolonged their treatment time. Only 60% of the 128 patients in arm B completed the scheduled XRT as specified. The rest required more treatment breaks than prescribed. The mean total treatment time was 51 days (10 ? 80) for arm A and 53 days (44 ? 135) for arm B (p < 0.0001). While not a large difference in the mean, the values were statistically significance. It may have been more useful to reveal the total treatment time as a median.
CTX was given for the first cycle at full dosages to the 130 patients in arm B. In the second and third treatment courses, approximately 75% and 50% patients received full dosages, respectively. Acute toxicity was significantly worse in arm B than arm A as expected, but late toxicities were equivalent. The major acute toxicities were severe mucositis and dermatitis. The CTX cause severe hematologic toxicity with grade 3 to 4 leukopenia of 15%. Severe nausea and vomiting was reported in 11% of the patients on arm B.
Locoregional control was evaluated in those 259 patients that received at least 6800 rads. For the XRT alone arm, 34% had complete remission of cancer without relapse. For the combined XRT/CTX arm, 44% had complete response without recurrence. Kaplan-Meier estimates reveal a 3-year locoregional control of 35% in arm B versus 17% in arm A. Distant metastases rates are approximately 10% for both treatment arms. Sites of distant disease were lung, bone and brain. Overall survival at 49% at 3 years for arm B versus 24% for arm A.
This study is a rare entity demonstrating a survival advantage to chemotherapy concurrent with radiation therapy as opposed to radiation therapy alone in the sites of the head and neck including oral cavity to the pyriform sinuses. Chemotherapy's effectiveness has been established for nasopharynx cancers, but at other head and neck sites, most of the literature from Italy, Duke University and Cleveland Clinic has demonstrated a locoregional control benefit, but no true overall survival improvement.
The XRT alone arm faired poorly relative to the other studies with the same population. This may be due to the planned breaks built into the treatment program. There has been a significant amount of literature from University of Florida demonstrating a 15% detriment in local control in patients that received split course XRT versus continuous course XRT.
Another difficulty of the article is that there was no separation of the data based on site. Oral cavity cancers do not run the same course as oropharynx malignancies, which are different than hypopharynx carcinomas, etc... It may have been that certain types of head and neck cancers are benefited more by the addition of chemotherapy, but that data are available. Median follow up was short as evidenced by figure 3, which shows those patients at risk. Perhaps longer follow up will help us determine whether the survival benefit is real or an initial benefit.
No mention of the salvage regimen (i.e. surgery, more XRT and/or CTX) utilized when these patients failed since a great majority of them did. Most of these patients were "unresectable." Thus, did the patients respond so well that salvage surgery was possible? The authors did not postulate as to why the survival was so much better than the local control rates in arm B relative to arm A.
Overall, CTX along with XRT for advanced head and neck cancers is advantageous in regards to local control. Survival advantage has been demonstrated in this German trial, but conclusions must be made with some reservations based criticisms we mentioned above.