Patient-reported outcomes from POINTBREAK: The randomized, open-label, phase III study of pemetrexed (pem) + carboplatin (cb) + bevacizumab (bev) followed by maintenance pem + bev versus paclitaxel (pac) + cb + bev followed by maintenance bev in patients with stage IIIb or IV nonsquamous non-small cell lung cancer (NS-NSCLC).
53 Background: Treatment (tx) impact on pt quality of life (QoL) informs pt and physician choices in the management of NSCLC. QoL, a secondary endpoint of POINTBREAK, was assessed using the patient-reported Functional Assessment of Cancer Therapy General (FACT-G), Lung (FACT-L), and Neurotoxicity (FACT&GOG-Ntx). Methods: Pts with previously untreated stage IIIB/IV NS-NSCLC and ECOG performance status 0-1 were randomized to receive induction Pem+Cb+Bev or Pac+Cb+Bev for up to 4 cycles. Pts without progressive disease received maintenance Pem+Bev (Pem Arm) or Bev (Pac Arm). Pts completed the FACT-G [physical, emotional, functional, and social well-being subscales (SS)], FACT-L (FACT-G + lung cancer SS), and FACT&GOG-Ntx (FACT-G + Ntx SS) at each visit. SS score and total score (TS) were calculated for each instrument. Tx group differences were analyzed using Linear Mixed Effects model. Toxicity was investigator assessed using CTCAE and was compared with Fisher’s exact test. Results: The primary endpoint of superior OS for the Pem arm was not met and is reported elsewhere. Pt overall compliance on the FACT was 91.2% (Pem Arm) and 90.0% (Pac Arm). For both the FACT-G and FACT-L TS, no overall significant change from baseline difference was observed between the Pem and Pac Arms. No overall differences were seen in SS scores except for Ntx. Pts on the Pem Arm had significantly less change from baseline in Ntx SS score and Ntx TS than pts in the Pac Arm at cycle 2 (p<0.001), and this significant tx effect persisted through induction and the six cycles of maintenance considered in the analysis. Significantly fewer pts on the Pem Arm experienced investigator assessed Grade 3/4 sensory neuropathy compared to the Pac Arm (0.0% vs. 4.1%, p<.00001). Conclusions: Based on the FACT results, pts on the Pem and Pac Arms reported similar changes in QoL except for less change from baseline in Ntx on the Pem Arm. Patient-reported Ntx is consistent with the clinically-measured drug-related Grade 3/4 sensory neuropathy.