Adding Chemo to RT of No Benefit in High-Risk H&N Cancer

NEW ORLEANS—The addition of concurrent cisplatin (Platinol) chemotherapy to radiation therapy after surgery failed to significantly improve locoregional control of high-risk head and neck cancers in the RTOG 9501/Intergroup phase III trial reported at the plenary session of the American Society for Therapeutic Radiology and Oncology (abstract plenary 3).

"We are disappointed to learn that despite added toxicity from the addition of chemotherapy, locoregional control, distant control, overall survival, and disease-free survival were not significantly improved," said Jay S. Cooper, MD, of New York University Medical Center.

Locoregional recurrence of disease has been the most common mode of failure in advanced head and neck cancer, despite grossly or microscopically complete surgical resection and postoperative radiation therapy, Dr. Cooper noted.

Between 1995 and 2000, this prospectively randomized phase III study enrolled 459 patients with resected, high-risk squamous cell carcinomas of the head and neck region. Subjects were considered high-risk by having two or more involved lymph nodes, extracapsular disease, and/or microscopically involved mucosal margins of resection.

Patients were randomized after resection to radiotherapy alone (60 to 66 Gy in 30 to 33 fractions over about 6 weeks) or identical radiotherapy plus cisplatin 100 mg/m2 on days 1, 22, and 43. Sixty-one percent of patients received all three doses of chemotherapy.

Study Results

At 37 months median follow-up of 415 patients, the addition of cisplatin provided no statistically significant improvement over radiotherapy alone in preventing locoregional or distant recurrence or in enhancing overall survival, Dr. Cooper reported.

The 2-year locoregional recurrence rate was 29% for radiotherapy alone and 24% for radiation plus chemotherapy, producing similar survival curves.

Overall survival at 2 years was 57% for radiotherapy and 65% for chemo-radiation, and disease-free survival was 43% vs 52%.

Locoregional recurrence as the sole site of first treatment failure occurred in 21% of the radiotherapy group and 16% of the concurrent combination therapy group. Distant metastases as the sole site of treatment failure occurred in 23% and 19%, respectively. These differences were not statistically significant. In terms of the patterns of failure, "the trends go in the direction we would like, but the overall P value is still 0.12," he added.

On the other hand, there was a significant difference in total recurrences, with 58% of radiotherapy patients having a recurrence of any type, compared with 45% in the chemoradiation group (P = .02), he reported.

Toxicity

Toxicity was worse in the chemora-diation arm, with grade 3 toxicity occurring in 51%, grade 4 in 24%, and grade 5 in 2%. For radiotherapy alone, grade 3 toxicity was seen in 40% and grade 4 in 6% (no grade 5). Deaths attributed to treatment occurred in 2% of patients undergoing chemoradiation but in none of the radiotherapy-alone patients.

"When we sum everything up and ask whether radiation and chemotherapy was a success or a failure, we can see that there are differences in the way tumors behave after radiation and chemotherapy but they are not statistically significant," Dr. Cooper said. "And one pays the price in toxicity; there is a highly significant increase in serious adverse events when you add chemotherapy."

Helmuth Goepfert, MD, professor and chairman of Head and Neck Surgery, M.D. Anderson Cancer Center, commented that this trial may have included too-few truly high-risk patients due to a disproportionate number of patients with cancer of the oropharynx as the primary site. "If you just treat the oropharynx with radiotherapy alone, you obtain a good result, so this group had basically a good prognosis," he noted. Overall, Dr. Goepfertr said, the results suggest that "chemotherapy in this group of patients only delays the inevitable."

On the bright side, Dr. Cooper said, "we did demonstrate that we can reliably identify this aggressive group of tumors from features seen on pathology exam, which should improve selection of patients for future trials and treatments."

The study also showed, he said, that "with modern techniques and meticulous attention to the details of our gold standard treatment—surgery followed by radiation therapy—even for these high-risk patients, we reduced the rate of locoregional recurrence to the point that only one of four patients have treatment fail them in this fashion."

The significant difference observed for treatment failure of any type suggests that "we are headed in the right direction," Dr. Cooper said. "Now we need to identify more effective agents."