DUBLINThe question of whether patients with locally advanced non-small-cell lung cancer (NSCLC) would live longer if they underwent chemotherapy prior to surgical resection would appear to be yes, based on two randomized trials reported in 1994.
More strongly debated are the relative advantages of surgery versus radiotherapy as locoregional treatment following induction chemotherapy. Other researchers are suggesting that by giving induction chemotherapy concurrently with radiotherapy, they may be able to avoid surgery and instead simply continue the radiotherapy to a specified total dose. These issues were on the table at the 8th World Conference on Lung Cancer.
A recently completed trial from the Radiation Therapy Oncology Group (RTOG) has failed to tip the scales in favor of either surgery or radiation after induction chemotherapy, said Valerie W. Rusch, MD, of the Thoracic Service at The Memorial-Sloan Kettering Cancer Center. She reported that, because of poor accrual, the RTOG 89-01 phase III trial had to be discontinued after only 75 patients had been randomized. Eligible patients had operable NSCLC with spread to the mediastinal lymph nodes (N2 disease), were medically able to receive cisplatin, and had sufficient cardiopulmonary reserve to undergo resection.
Participants first received induction chemotherapy consisting of two cycles of cisplatin (120 mg/m2 on days 1 and 29) and vinblastine (4.5 mg/m2 on days 1, 15, 29, and 43) with or without mitomycin (8 mg/m2 on days 1 and 29).
They were then randomly assigned to either surgery with complete ipsilateral node dissection or radiotherapy to a total dose of 50 Gy given in 2-Gy fractions to an enlarged field plus a 14-Gy boost to the area of gross disease. Locoregional treatment was followed by consolidation chemotherapy with two cycles of cisplatin plus vinblastine.
Dr. Rusch noted that nearly 80% of patients were able to complete induction chemotherapy and that only seven patients experienced disease progression during chemotherapy. Surgery was performed in 90% of eligible patients, only one quarter of whom had a pathologically incomplete resection; similarly, radiotherapy was completed in 87% of cases.
Both multimodality approaches were generally well tolerated, yielding equivalent rates of treatment-related morbidity and chemotherapy-associated toxicity.
Two thirds of patients in both groups were alive at one year, and median survival reached 19.3 months in the surgery group and 17.4 months in the radiotherapy group. With these relatively small numbers of randomized patients, no differences in overall or median survival between the surgery and radiation study arms were observed, she said.
Dr. Rusch noted that the feasibility of these two treatment approaches and the lack of observed survival differences between the two study arms indicate that a larger definitive trial testing this concept is warranted.
One such large randomized trial is taking a somewhat different approach. The RTOG is now coordinating the North American Inter-group 0139 trial, which has thus far enrolled about 200 patients with pathologically documented N2 disease and is striving for a total sample size of 510 patients.
This trial is testing whether surgical resection following induction chemotherapy plus radiation is superior to the same induction chemotherapy plus radiation regimen followed by completion of a definitive radiotherapy dose.
After two cycles of cisplatin-etoposide coupled with concurrent radiotherapy, Intergroup patients are randomly assigned to undergo surgery or to continue radiotherapy without interruption to a total dose of 61 Gy. Both treatment arms include two cycles of consolidation therapy with cisplatin and etoposide.
Intergroup principal investigator Kathy Albain, MD, of Cardinal Bernadin Cancer Center of Loyola University, Maywood, Ill, described divergent patterns of nonhematologic toxicity in the two treatment arms, based on preliminary data from 132 patients.
To date, there have been no toxic deaths in the radiotherapy arm despite such grade 4 complications as esophagitis and pneumonitis. In contrast, there have been two postoperative deaths attributed to ARDS (acute respiratory distress syndrome) and two others due to cardiopulmonary causes.
Accrual rates for this trial have increased throughout North America, Dr. Albain said. In her talk, she addressed earlier critiques of the study design: We now know that toxicity is not prohibitive, and that both arms are safe in a broad-based cooperative group setting. We also know from two recent trials that concurrent chemoradiation yields encouraging results. First, she said, a Japanese trial demonstrated the superiority of concurrent chemoradiation versus chemotherapy followed by radiation (Furuse K et al: ASCO 1997, abstract 1649see page 11 for an update of this trial).
Second, follow-up survival data from two small pilot trials from the Southwest Oncology Group (SWOG) have shown, at least in patients with stage IIIB disease, similar survival results with chemotherapy plus radiation and with chemotherapy plus radiation followed by surgery.
This gives us some additional confidence that we can proceed with the accrual in our current N2 disease trial to answer this important question, Dr. Albain said.
A complementary study from Europe, the five-country, 32-center EORTC 08941 trial, is predicated on the assumption that neoadjuvant chemotherapy plus surgery will provide a 10% five-year survival edge over chemotherapy plus radiotherapy, said investigator Ted Splinter, MD, of University Hospital Dijkzigt, Rotterdam, The Netherlands. The EORTC trialists have already entered more than 40% of their planned 406 patients with stage IIIa (N2) inoperable NSCLC.
In a preliminary analysis of data from 124 patients, two thirds showed a major response to three courses of cisplatin- or carboplatin-based induction chemotherapy and could thus be randomized to either surgery or radiotherapy at a total dose of 54 Gy.
The selection of responders for randomization was based on the premise that inoperable patients remain inoperable unless they achieve a major response to chemotherapy. The remaining nonre-sponders were treated with radiotherapy.
Pathologic downsizing as compared to clinical stage was achieved in 42% of patients, which is exactly what we want, Dr. Splinter told Oncology News International in an interview.
The EORTCs choice of response to chemotherapy as the criterion for randomization was vindicated by the finding that radical resection was possible in 89% of responders who were assigned to surgery. The predictive value of selecting responders was excellent, Dr. Splinter commented. It is the perfect selection criterion.
Thus far in this trial, there have been two toxic deaths during induction chemotherapy and one postoperative death.
I think prospective analysis will indicate that recovery takes months longer following radiotherapy and that quality of life will probably be better following surgery, he said. So even if a 10% survival improvement is not found, the quality of life advantage should make the difference in determining the new standard for combined modality therapy.
He expressed the hope that a combined analysis of results from the EORTC, Intergroup, and RTOG trials could yield answers supported by a 1,000-patient-strong data base.