Background: A partial response according to the Response Evaluation Criteria in Solid Tumors includes a wide range of changes in tumor size. This study evaluated whether further specification of tumor reduction based on the depth of response (DpR) would provide a more precise association with outcomes for patients with non–small cell lung cancer (NSCLC) treated with first-line platinum-based chemotherapy. Methods: A retrospective analysis was performed for the randomized phase 3 CA031 trial in patients with NSCLC treated with carboplatin in combination with nab-paclitaxel or solvent-based paclitaxel. Quartiles according to the maximum tumor reduction from the baseline were defined (quartile 1 [Q1], >0% to 25%; quartile 2 [Q2], >25% to 50%; quartile 3 [Q3], >50% to 75%; and quartile 4 [Q4], >75%) and were compared with those patients with no tumor reduction (NTR). The primary objective was to evaluate the association between DpR and overall survival (OS). Results: Of the 1052 patients enrolled in the CA031 trial, 959 (91%) were evaluable, and they included 365 (38.1%) who were classified as Q1, 327 (34.1%) who were classified as Q2, 131 (13.7%) who were classified as Q3, and 34 (3.5%) who were classified as Q4; 102 had NTR (10.6%). The median OS values for patients in the NTR, Q1, Q2, Q3, and Q4 groups were 4.8, 10.4, 14.5, 19.3, and 23.5 months, respectively. The maximum DpR on treatment was an independent predictor of improved OS in comparison with patients with NTR; the hazard ratio decreased from 0.43 in Q1 to 0.16 in Q4. Conclusions: DpR was strongly associated with OS in patients with NSCLC receiving first-line platinum-based therapy. Additional studies may help to define the role of DpR in solid tumors.