Favezelimab Combo Boosts PFS vs Chemo in PD-1–Refractory Hodgkin Lymphoma

Favezelimab, a LAG-3 inhibitor, plus pembrolizumab (Keytruda) exhibited a significant progression-free survival (PFS) benefit and trended toward favorable overall survival (OS), but with a lower objective response rate (ORR), vs chemotherapy among patients with relapsed/refractory classic Hodgkin lymphoma (cHL), according to findings from the phase 2 KEYFORM-008 trial (NCT05508867) presented at the European Hematology Association (EHA) 2026 Congress.

In the intent-to-treat (ITT) population, the median PFS in the favezelimab/pembrolizumab arm (n = 104) was 6.4 months (95% CI, 5.7-8.4) vs 5.7 months (95% CI, 5.6-6.2) in the chemotherapy arm (n = 99; HR, 0.757; 90% CI, 0.575-0.996; P = .0462). The respective 6-, 9-, and 12-month PFS rates were 50.7% vs 45.5%, 38.3% vs 22.8%, and 31.8% vs 13.8%.

A subgroup analysis for PFS revealed that the favezelimab-containing regimen conferred particular benefit to patients who received prior autologous stem cell transplantation (ASCT; HR, 0.54; 95% CI, 0.34-0.84), as well as for patients who were younger than 65 years (HR, 0.68; 95% CI, 0.47-0.98). Moreover, those who did not receive PD-1 inhibition as the last line of therapy experienced a pronounced benefit with the LAG-3 inhibitor (HR, 0.50; 95% CI, 0.26-0.95).

Regarding median OS, it was not reached in either arm, but data showed a favorable trend for favezelimab/pembrolizumab (HR, 0.67; 95% CI, 0.33-1.36). Moreover, the 6-, 9-, and 12-month OS rates in the investigational and chemotherapy arms were 98.1% vs 94.9%, 93.2% vs 91.8%, and 92.0% vs 88.0%.

The ORR was higher among the chemotherapy arm; the rates in the favezelimab-containing and chemotherapy arms were 44.2% (95% CI, 34.5%-54.3%) vs 67.7% (95% CI, 57.5%-76.7%). Additionally, the complete remission (CR) rate was 13.5% vs 36.4%, respectively, with 30.8% vs 31.3% of each experiencing partial responses. Moreover, the disease control rate (DCR) was similar in each arm; the respective rates were 67.3% (95% CI, 57.4%-76.2%) vs 71.7% (95% CI, 61.8%-80.3%).

The median duration of response (DOR) per Lugano criteria and investigator assessment was 11.0 months (range, 0.0+ to 21.8+) with favezelimab/pembrolizumab vs 5.4 months (range, 0.0+ to 27.6+) with chemotherapy. The 6-, 9-, and 12-month DOR rates were 63.5% vs 31.5%, 55.0% vs 20.5%, and 42.9% vs 20.5% in each arm, respectively.

Finally, among patients who crossed over from chemotherapy to favezelimab/pembrolizumab (n = 52), the median PFS was 5.5 months (95% CI, 2.9-8.3), with respective 6-, 9-, and 12-months rates of 42.2%, 26.5%, and 20.6%. Median OS was not reached for this population, with 6-, 9-, and 12-month rates of 100%, 97.3%, and 87.1%. The ORR was 42.3% (95% CI, 28.7%-56.8%) for this population, with a CR rate of 9.6%.

“The trial met its primary end point. [Favezelimab/pembrolizumab] showed significant PFS benefit with early separation of curves at 6 months, although the small number of [patients] at risk thereafter warrants cautious interpretation,” Guilherme Fleury Perini, MD, hematologic oncologist specializing in lymphoma at the Hospital Israelita Albert Einstein in São Paulo, Brazil, stated in the presentation. “The results suggest that LAG-3 inhibition in combination with PD-1 blockade may provide clinical activity in a very heavily pretreated PD-1–refractory population.”

Perini noted that investigators discontinued the KEYFORM-008 trial based on the sponsor’s decision to terminate the entire favezelimab clinical development program.

Patients 18 years and older with PD-1–refractory relapsed/refractory cHL were randomly assigned 1:1 to receive favezelimab/pembrolizumab or chemotherapy. Additionally, those enrolled had an ECOG performance score of 0 to 2, disease progression within 12 weeks of last anti–PD-1 therapy, and ineligibility for ASCT or brentuximab vedotin (Adcetris).

In the investigational arm, patients received 800 mg of favezelimab and 200 mg of pembrolizumab intravenously every 3 weeks for up to 35 cycles. Chemotherapy consisted of intravenous bendamustine at 90 to 120 mg/m² on days 1 and 2 of each 21- or 28-day cycle or intravenous gemcitabine at 800 to 1200 mg/m² on days 1 and 8 of each 21-day cycle. Disease progression was confirmed by blinded independent committee review (BICR), and crossover to the investigational arm was permitted if progression occurred with chemotherapy.

The median age of patients in the investigational and chemotherapy arms was 40.5 years (range, 18-84) and 40.0 years (range, 18-82), respectively. Most patients in each arm were male (55.8% vs 60.6%), White (78.8% vs 77.8%), and had an ECOG performance status of 0 (63.5% vs 59.6%). Additionally, 70.2% vs 69.7% had refractory disease status and 48.1% vs 53.5% received prior ASCT. The median number of prior lines in each arm were 5 (range, 2-12) and 5 (range, 2-13), respectively; 90.4% vs 89.9% received prior brentuximab vedotin, 76.9% vs 75.8% received an anti–PD-1 as most recent therapy, and 36.5% vs 39.4% received prior oncologic radiation.

The primary end point of the trial was PFS per Lugano criteria by investigator assessment. Secondary end points included investigator-assessed ORR and DOR per Lugano criteria, as well as OS and safety and tolerability.

In the favezelimab/pembrolizumab and chemotherapy arms, adverse effects (AEs) occurred in 97.1% vs 96.9% of each. Grade 3 to 5 AEs occurred in 28.8% vs 50.5%; AEs led to discontinuation of treatment in 14.4% vs 4.1% and death in 1.9% vs 3.1%. Treatment-related AEs (TRAEs) occurred in 76.0% vs 83.5% of each arm, of which 18.3% vs 40.2% were grade 3 to 5; 12.5% vs 3.1% lead to treatment discontinuation, and 0% vs 1.0% led to death. Immune-mediated AEs or infusion reactions were reported in 29.8% vs 13.4% of each arm, of which 4.8% and 5.2% were grade 3 or 4 in severity; 2.9% in the favezelimab/pembrolizumab arm led to treatment discontinuation.

The most common TRAE in the investigational arm was hypothyroidism in 15.4% of patients. In the chemotherapy arm, nausea (39.2%), anemia (25.8%), fatigue (17.5%), neutrophil count decreases (17.5%), and platelet count decreases (17.5%) were among common TRAEs. Moreover, neutrophil count decreases, lymphocyte count decreases, and anemia were among the most common grade 3 or 4 TRAEs.

In the crossover arm, any-grade AEs occurred in 82.7% of patients, of which 28.8% were grade 3 to 5. A total of 11.5% of patients experienced AEs leading to discontinuation, and a further 3.8% had died due to AEs. The most common AEs in this group included diarrhea (13.5%), hypothyroidism (11.5%), pyrexia (11.5%), and alanine aminotransferase increases (11.5%).

Disclosure: Perini declared advisory/consultancy roles for AbbVie, AstraZeneca, BeOne, and Johnson & Johnson; speaker bureau roles for AbbVie, AstraZeneca, BeOne, and Merck Sharp & Dohme (MSD); as well as travel, accommodation, and other expenses funded by AbbVie, AstraZeneca, and Roche.

Reference

Perini GF, Lavie D, Bröckelmann PJ, et al. Coformulated favezelimab/pembrolizumab versus chemotherapy in participants with relapsed or refractory classic Hodgkin lymphoma: results of the randomized phase 2 KEYFORM-008 study. Presented at: European Hematology Association 2026 Congress; June 11-14, 2026; Stockholm, Sweden. Abstract S220.