Abstract
BACKGROUND: The National Cancer Institute's Common Terminology Criteria for Adverse Events (CTCAE) and adjustment rules after severe toxicity are derived by consensus, but to the authors' knowledge little is known regarding the determinants of toxicity recurrence, especially in the elderly. METHODS: The authors prospectively accrued 200 patients (aged ≥65 years) before chemotherapy. For those with CTCAE grade 3 to 4 nonhematologic or CTCAE grade 4 hematologic toxicities (severe toxicity), the duration and functional impact, treatment modifications, and severe toxicity recurrence were recorded. The regimen's toxicity was adjusted with the MAX2 index, the average of the most frequent grade 4 hematologic toxicities and the most frequent grade 3 to 4 nonhematologic toxicities reported in publications of a regimen. RESULTS: The median patient age was 73 years (range, 65-90 years). Among 163 patients who were evaluable for toxicity after ≥1 treatment cycle (receiving on average 4.73 cycles), 82 had severe toxicity, 10 were discontinued for toxicity, 6 were discontinued for other reasons, and 5 patients had died. Sixty-one patients received further chemotherapy: 41 without dose modification (16 with secondary prevention measures) and 20 with dose modifications. Without modification, 19 patients (46%) experienced toxicity recurrence (0 deaths). With modification, 7 patients (35%) experienced a toxicity recurrence (1 death). On univariate analysis, treatment intent, hospitalization, duration-adjusted activities of daily living (ADL), quality of life impact, and fatigue were associated with dose modification. ADL remained associated on multivariate analysis (P = .02). On univariate analysis for toxicity recurrence, Eastern Cooperative Oncology Group performance status and MAX2 score demonstrated an association, with only the latter found to remain statistically significant on multivariate analysis (P = .04). CONCLUSIONS: If a severe toxicity does not have a long duration of impact on ADL, oncologists are less inclined to modify treatment. With proper supportive measures, this leads to recurrence risks similar to those shown in patients with modified treatment, with low risks of toxic deaths overall.