Source: www.oncologyreport.com
Author: Miriam E. Tucker

Adding chemotherapy to radiotherapy improved 10-year survival of resectable head and neck carcinomas among high-risk patients who had microscopically involved resection margins and/or extracapsular spread of disease – but not in high-risk patients who only had tumor in multiple lymph nodes.

The findings come from a long-term update and unplanned subset analysis of 410 evaluable patients from the RTOG (Radiation Therapy Oncology Group) 9501 phase III study, which previously showed no overall survival advantage from the addition of cisplatin chemotherapy to radiation.

The new data are “good news,” according to lead author Dr. Jay Cooper, director of Maimonides Cancer Center in Brooklyn, N.Y.

“We now can eradicate some advanced head and neck tumors that we couldn’t before by adding chemotherapy to radiation therapy. At the same time, we can spare other patients who would not do better with the addition of chemotherapy from its side effects,” he said at a head and neck cancer symposium sponsored by the American Society for Radiation Therapy.

The RTOG 9501 study randomized 459 patients with high-risk, resected head and neck cancers to receive either radiation therapy of 60 Gy in 6 weeks (RT), or identical radiotherapy plus cisplatin at 100 mg/m2 IV on days 1, 22, and 43 (RT+CT).

When reported at a median follow-up of 45.9 months, the locoregional control rate was significantly higher in the combined-therapy group than in the group given radiotherapy alone (hazard ratio for locoregional recurrence, 0.61); disease-free survival was significantly longer with combined therapy (HR for disease or death, 0.78), but overall survival was not (HR for death, 0.84). Moreover, four patients who received combination therapy died as a result of treatment (N. Engl. J. Med. 2004;350:1937-44).

In the current updated analysis conducted 10 years post treatment, none of the primary outcomes differed significantly between the two groups. The evaluable population comprised 208 patients who received RT and 202 given RT+CT. For patients treated by RT vs. RT+CT, the rates were, respectively, 28.8% vs. 22.3% (P = .10) for locoregional failure, 19.1% vs. 20.1% (P = .25) for disease-free survival, and 27.0% vs. 29.1% (P = .31) for overall survival.

In a subset analysis that had not been performed previously, however, statistically significant differences appeared within the 242 patients who had microscopically involved resection margins and/or extracapsular spread of disease. In this group, 115 patients received RT and 127 were given RT+CT. Locoregional failure occurred in 33.1% of the CT group vs. 21.0% of those treated with RT+CT (P = .02). The disease-free survival rate was 12.3% vs. 18.4% (P = .05), and the overall survival rate was 19.6% vs. 27.1% (P = .07), with both end points favoring RT+CT.

That left 168 patients who did not have involved margins or extracapsular extension and were included in the trial solely because they had multiple involved nodes. In this group, 93 received RT and 75 RT+CT, with no significant difference in long-term outcomes.

There was a trend in improved cause-specific survival with RT+CT for patients whose death resulted from head and neck cancer, but more deaths not due to the study cancer were observed in patients treated with concurrent cisplatin. This is a hypothesis-generating finding that needs to be investigated in future trials, Dr. Cooper noted.

In an interview, he explained that the rationale for looking specifically at the patients with microscopically involved resection margins and/or extracapsular spread came from a previous analysis of the raw data from the RTOG trial combined with those of a concurrently published study conducted by the EORTC (European Organisation for Research and Treatment of Cancer).

Although the design of the EORTC 22931 study was similar, the outcome was different. Of a total 167 patients who had been randomized to RT or RT+CT, the rate of progression-free survival at a median follow-up of 60 months was significantly higher in the combined-therapy group than in the group given radiotherapy alone (P = .04) (N. Engl. J. Med. 2004;350:1945-52).

When the data from RTOG and EORTC were combined, it became clear that the main reason for the difference in outcome was in the different entry criteria for the two trials, and that extracapsular extension (ECE) and/or microscopically involved surgical margins were the only risk factors for which the impact of adjuvant chemotherapy-enhanced radiation therapy was significant in both trials (Head Neck 2005;27:843-50).

“What we learned from that analysis is that the patients who got on one of the trials but wouldn’t have qualified for the other trial were not getting much benefit from the study regimen, whereas those who qualified for either study – due to having involved margins and/or extracapsular extension – did better with chemotherapy on all three measures.

“These results were highly significant. But more importantly, the data suggested a subgroup where the big bang for the buck was,” Dr. Cooper said in the interview.

The findings don’t mean that the patients who did not benefit are not “high risk,” Dr. Cooper said. “Would they benefit from other chemotherapy? We don’t know. Would they benefit from different drugs or different regimens? Maybe. But we can now fairly comfortably say that in both the short and long run, if patients are high risk only because of involved lymph nodes, don’t treat them with this combination, and spare them the toxicity.”

Dr. Cooper stated that he has no disclosures, as did nine coauthors. One additional coauthor is an employee of Lilly USA.