Head-to-head comparison of pembrolizumab-axitinib versus pembrolizumab-lenvatinib in metastatic RCC: Real world data from the Canadian Kidney Cancer Information System (CKCis) database.

460 Background: Immune checkpoint inhibitor-tyrosine kinase inhibitor (IO-TKI) combinations have transformed first-line treatment for metastatic renal cell carcinoma (mRCC). Pembrolizumab-axitinib (PA) and pembrolizumab-lenvatinib (PL) have each demonstrated efficacy in randomized trials, but real-world comparative data remain limited. Methods: We conducted a retrospective cohort study using the CKCis registry. Patients with confirmed mRCC who received first-line PA or PL were included. Primary endpoint was progression-free survival (PFS). Key secondary endpoints included overall survival (OS) and safety. Results: A total of 632 patients were identified (median age 66; 73.5% male), including 516 (81.6%) with clear cell and 116 (18.4%) with non-clear cell histology. Of these, 549 (86.9%) received PA and 83 (13.1%) received PL. Median follow-up was 28.9 months. Baseline characteristics were generally comparable between groups, although differences were observed in histology, performance status and the distribution of brain and lung metastases. At 12 months, the PFS rate was higher with PL compared with PA (75.9% vs. 57.1%); 5-year PFS was 36.2% vs. 25.9%, respectively (P < 0.01). 12-month OS was 93.9% for PL and 85.2% for PA; 5-year OS was 65.9% vs. 51.4% (P = 0.0172). For non-clear cell histology, 12-month OS was 91.6% for PL and 78% for PA (P = 0.0335). Among clear cell histology, 12-month OS was 94.9% for PL and 86.6% for PA (P = 0.0884). In the IMDC favourable-risk group, 12-month OS rate was 100% with PL and 95.2% with PA (P = 0.5994). For intermediate/poor IMDC risk, 12-month OS rate was 92% with PL and 82.3% with PA (P = 0.0145). Multivariable Cox analysis adjusting for age and IMDC risk showed that PA was associated with an increased risk of death compared with PL (HR = 2.15; 95% CI, 1.14-4.06; P = 0.019). IMDC risk was an independent predictor of OS, with patients in the intermediate/poor-risk group demonstrating a 94% higher hazard of death relative to those in the favourable-risk group (HR = 1.94; 95% CI, 1.32–2.87; P = 0.001). Dose reductions of the TKI occurred more frequent with PL (50%) than with PA (35.3%). Both regimens were associated with manageable but frequent toxicities. Fatigue, diarrhea, and anorexia were commonly observed with PA. PL showed a similar overall profile but a higher rate of Grade 3 proteinuria. No Grade 5 events occurred. Conclusions: In this real-world mRCC cohort, treatment with PL was associated with better PFS and OS compared to PA, but higher dose modification rates. These findings align with outcomes observed in pivotal trials and suggest that PL may offer a clinically meaningful benefit in routine practice. Prospective studies are needed to validate these results and to further explore the balance between efficacy and tolerability among first-line IO-TKI combinations.