Treatment of small cell lung cancer in the elderly

Since both the incidence of lung cancer and the proportion of the population over age 65 are increasing rapidly in North America, we undertook a retrospective review of elderly patients with small cell lung cancer (SCLC) in an attempt to assess the effect of age on treatment decisions, response, survival, and toxicity.

Retrospective chart view.

Oncology Unit of a university-affiliated teaching hospital.

There were 123 patients age > 70 years treated from 1976-88. Chemotherapy consisted of either cyclophosphamide, doxorubicin, and vincristine, or etoposide and cisplatin.

There were 74 patients aged 70-74, 35 aged 75-80, and 14 aged 80 years or older. No significant differences existed between the groups in sex, stage, performance status, or presence of co-morbid disease. Median survivals for patients with limited and extensive disease were 11.9 and 5.2 months, respectively (P = < 0.0001), with no significant difference for patients in any age group (P = 0.4). For both limited and extensive disease, survival correlated strongly with the treatment received. Twenty-five patients received no treatment (median survival 1.1 months), 20 had radiation only (median 7.8 months), and 27 patients had < 3 cycles of chemotherapy (median 3.9 months). Median survival for the 50 patients who had 4-6 cycles was 10.7 months (limited disease 15.0 months, extensive disease 8.61 months). In the Cox Model, survival correlated strongly with stage of disease and chemotherapy treatment (P < 0.0001), but only marginally with performance status (P < 0.077). Of the 77 patients who had chemotherapy, less than 50% in all age groups completed six cycles. Only two patients completed chemotherapy without a single dose reduction, and 76.7% required more than two reductions.

Chemotherapy should not be withheld from elderly patients with SCLC on the basis of age. The survival of patients who receive chemotherapy is significantly longer than that of untreated patients even though frequent dose reductions for toxicity may be required. The survival benefit is due to treatment effect and is not due to a selection bias in the cohort of patients chosen for therapy.