The American Cancer Society estimated that 180,000 patients in the United States would develop lung cancer during 1996. Approximately 80% of these patients were expected to have non-small-cell lung cancer. In approximately one-third of cases, the disease is potentially resectable for cure at the time of presentation. Another third of patients have locally advanced disease (stages IIIA and IIIB) that is not surgically approachable de novo with curative intent. The remaining one-third have stage IV disease, implying hematogenous dissemination at the outset. This article describes how management strategies for the latter two groups have evolved during the past 5 years.
Patients who have stage III non-small-cell lung cancer by virtue of T3 disease, but with N0 or N1 disease, may be appropriately considered for resection. This is contingent on their having adequate pulmonary function and an expected 5-year survival rate similar to that associated with stage II disease (ie, 30% to 40% of patients).
On the other hand, N2 disease, regardless of tumor status, is generally a contraindication to initial surgical resection; the exception is when only one nodal area (or station) in the mediastinum is involved. This type of presentation is seen in a small, favorable subset of patients who exhibit a 5-year survival rate of 25% to 30% after complete resection. (Most also receive postoperative irradiation.)
Sequential Combined-Modality Therapy
Historically, patients with multiple sites of N2 disease (stage IIIA) and all those with T4 or N3 disease (stage IIIB) have been treated with radiation therapy alone. In the United States, however, combined-modality therapy (chemotherapy plus irradiation) has become the new standard of care, based largely on outcome data from two prospective, randomized trials.
The first trial, reported by Dillman et al for the Cancer and Leukemia Group B (CALGB), compared standard continuous, daily fractionated irradiation (to 60 Gy in 6 weeks) with a regimen in which two cycles of vinblastine(Drug information on vinblastine) and cisplatin(Drug information on cisplatin) (Platinol) preceded the same radiation therapy. Compared with irradiation alone, the combined-modality approach was associated with a statistically significant survival advantage (median duration, 9 vs 14 months; 2-year rate, 13% vs 26%, respectively; P < .012). The survival difference persisted at 5 years (7% vs 19%, respectively).
Unfortunately, entry in this trial was closed after 155 patients had been enrolled, before patient accrual reached the level that had been initially projected. Because of concerns that the results might represent a false-positive finding, a second, larger trial (involving 450 patients) was mounted as an intergroup study.
As reported by Sause et al, this study also demonstrated a significant survival advantage when two cycles of vinblastine and cisplatin were given before continuous, fractionated irradiation to 60 Gy (the CALGB arm) as opposed to either of two programs of radiation therapy alone. The 2-year survival rate was 32% with the combined-modality approach vs 19% with standard irradiation alone. Therefore, the combination of vinblastine and cisplatin followed by chest irradiation has been validated in consecutive randomized trials and has been widely adopted as a standard regimen.
Concurrent Combined-Modality Therapy
The Southwest Oncology Group (SWOG) pursued a different approach to combined-modality therapy in patients with stage III, inoperable non-small-cell lung cancer. These investigators chose to administer cisplatin and etoposide(Drug information on etoposide) (VePesid) concurrently with chest irradiation (two cycles of chemotherapy plus 45 Gy of continuous, fractionated local treatment), followed by surgical resection 3 to 5 weeks after completion of the combined-modality induction phase.[5,6]
The concurrent use of chemotherapy and irradiation was chosen because the SWOG had previously demonstrated its feasibility in the treatment of limited-stage small-cell lung cancer, and the results were apparently superior to those achieved when the modalities were administered in sequence. Another important distinction between this study and previous work was that before the initiation of any therapy, the investigators conducted surgical stagingincluding mediastinal nodal sampling, if necessaryto document N2 or N3 status. In the other studies, only clinical and radiographic criteria were used for staging.
Several important observations emerged from this phase II SWOG trial.[5,6] First, a comparable proportion of patients with stage IIIA N2 and stage IIIB disease could undergo resection (74% for patients with stage IIIA N2 disease and 58% for patients with stage IIIB disease). Second, survival rates were encouraging and were again comparable for stage IIIA and IIIB disease (median duration, 13 months for stage IIIA and 16 months for stage IIIB; 2-year rate, 34% for patients with stage IIIA disease and 38% for patients with stage IIIB disease). Third, almost 60% of the resected specimens contained no macroscopic evidence of residual tumor at the primary site, and the pathologic complete response rate was 20%.
The most striking clinicopathologic correlation, however, was between survival and mediastinal nodal status at the time of resection. Among 107 patients who underwent resection, 45% had microscopically negative mediastinal nodes; median survival duration in this group was 30 months and the 3-year survival rate was 40%. In contrast, patients with microscopically positive nodes had a median survival duration of 10 months and a 3-year survival rate of 10%. If these observations are borne out in other studies of neoadjuvant therapy, the implication will be that additional, potentially non-cross-resistant therapy may be of particular benefit in the latter group.
At face value, the results of the SWOG trial appear to be better than those for sequential, combined-modality therapy without surgery. However, the rate of treatment-related fatality was higher in the SWOG trial (10%, vs 2% for most other studies without surgery), and the majority of these deaths occurred in the postoperative setting.
Another trial of triple-modality therapy was reported recently by Strauss et al for CALGB. Like the SWOG trial, this study was confined to surgically staged patients. Unlike the SWOG trial, however, patients with stage IIIB disease were excluded and those with the more favorable stage IIIA disease (T3, N0; T3, N1) were included, comprising 20% of the participants.
Cisplatin, vinblastine, and fluorouracil(Drug information on fluorouracil) (5-FU) were combined with concurrent chest irradiation to 30 Gy, followed by resection, when possible, and then one more cycle of chemotherapy and an additional 30 Gy to the tumor bed. Of 41 eligible patients, 25 (61%) underwent resection. The median duration of survival was 16 months, and the 2-year and 5-year survival rates were 47% and 22%, respectively. The pathologic complete response rate was 16%.
As in the SWOG study, there was a high incidence of treatment-related deaths (15%), half of which were perioperative. It is noteworthy that patients who completed therapy according to the protocol but did not undergo resection had a long-term survival rate that was nearly identical to that in patients who were resected (25% vs 24%, respectively).
A recent SWOG trial assessed concurrent cisplatin and etoposide for four cycles plus chest irradiation to 61 Gy, omitting resection, in surgically staged IIIB disease. The findings revealed a 40% survival rate at 2 years and a reduction in mortality when compared with that observed in the trial incorporating surgery.
Taken together, these results suggest that a concurrent, combined-modality approach with full-dose chemotherapy (four cycles of cisplatin and etoposide) and definitive, continuous fractionated irradiation (60 Gy in 6 weeks) may be as good as the triple-modality approach with resection added. This hypothesis is the subject of an ongoing, randomized intergroup trial (INT 0139) in stage IIIA, N2 disease. If brought to successful completion, this study should resolve the question of whether surgery should play a major role in the treatment of stage III disease.
For stage IIIB disease, the SWOG has recently activated a trial of concurrent cisplatin plus etoposide (two cycles) with chest irradiation (60 Gy), followed by three cycles of docetaxel(Drug information on docetaxel) (Taxotere) consolidation, in an attempt to gain further improvement in long-term results. Another major question involves comparison of sequential vs concurrent chemoradiotherapy; this issue is the focus of a current study by the Radiation Therapy Oncology Group, as well as a recently completed trial in Japan. The results of these trials are awaited with interest.