In total, 206 patients were lesion-response-evaluable with available QoL data from baseline and ≥1 postbaseline visit and were included in the LCSS analysis. In general, symptom burden decreased and QoL improved during four cycles of treatment with nab-paclitaxel plus carboplatin. Positive changes from baseline were observed in several LCSS symptom scores, including cough, appetite, and pain. For the LCSS average total score, improvement from baseline peaked on cycle 2 day 1 and remained stable throughout the rest of the induction phase (Figure 1A). A similar trend was noted for the average symptom burden index (Figure 1B). For the lung cancer symptom score that rated the severity of disease-specific symptoms, improvements from baseline were observed during each of the four cycles of nab-paclitaxel plus carboplatin treatment (Figure 1C). The changes in global score remained generally stable with respect to baseline scores throughout the induction phase, as did the changes in the 3-item index scores (Figure 2A and B). Overall, 46% of patients achieved a clinically meaningful improvement (≥10 mm) from baseline in the composite score of pulmonary symptoms during the induction phase. Clinically meaningful improvements in the individual items of cough, shortness of breath, and hemoptysis were also reported in 58%, 49%, and 16% of patients, respectively.
Two-hundred and six patients who were lesion-response-evaluable with available QoL data from baseline and ≥1 postbaseline visit were also included in the EQ-5D-5L analysis. The majority of patients experienced stability or improvement from baseline in individual EQ-5D-5L items, including self-care (91%), mobility (88%), anxiety/depression (88%), pain/discomfort (85%), and usual activities (83%) (Table 2). In patients reporting problems at baseline, complete resolution from baseline at least once during four cycles of treatment was reported in 30% to 50% of patients in individual EQ-5D-5L dimensions.
(To view a larger version of Table 2, click here.)
LCSS and EQ-5D-5L by lesion response
Overall, of lesion-evaluable patients who completed baseline and ≥1 postbaseline QoL assessment, 110 were responders (CR/PR), and 96 were nonresponders. Responders had higher QoL scores compared with the overall population and nonresponders. LCSS average total score, average symptom burden index, and lung cancer symptom scores were improved from baseline in responders and were generally stable with respect to baseline in nonresponders (Figure 1A–C). Overall, 56% of responders and 33% of nonresponders achieved a clinically meaningful improvement in the composite score of pulmonary symptoms.
Across all EQ-5D-5L dimensions, stabilization or improvements in QoL scores were more pronounced in responders vs nonresponders and vs the overall population (Table 2). The percentage of responders reporting complete resolution at least once was nearly double that of nonresponders in the individual dimensions of usual activities (39% vs 21%), pain/discomfort (44% vs 27%), and anxiety/depression (59% vs 31%).