Serum Ferritin Predicts Treatment Response in Myelofibrosis for Those Assigned to Momelotinib

Serum ferritin at baseline was a predictor for week 24 transfusion independence response (TI-R) in patients with myelofibrosis receiving momelotinib according to findings from an analysis of data from the phase 3 SIMPLIFY trials presented in a poster at the 2021 ASH Annual Meeting and Exposition.¹

Stephen T. Oh, MD, PhD, an assistant professor of medicine and assistant professor of pathology and immunology in the Divisions of Hematology and Biology and Biomedical Sciences at Washington University School of Medicine, said these findings are further evidence supporting the differential activity of momelotinib and ruxolitinib in myelofibrosis.

“Serum ferritin, a well established and easily measured clinical biomarker, may be useful in treatment decision making in myelofibrosis, especially in patients with anemia and ferritin between 90 ng/mL to 650 ng/mL, in momelotinib demonstrates the greatest TI benefit over and ruxolitinib [Jakafi],” he said. “In patients with the highest ferritin levels, further optimization of therapy may be required to achieve effective TI responses, such as the use of combination regimens.”

In patients with baseline ferritin levels from 90 ng/mL to 650 ng/mL, week 24 TI-R rate was nearly double for those assigned to momelotinib compared with those assigned to ruxolitinib (72% vs 38%; OR, 4.21; 95% CI; 2.24-7.89; P = .0439).

Oh added that these results show that momelotinib is superior at addressing anemia, which may lead to improved overall survival (OS). Degree of anemia and transfusion dependence are 2 of the most important predictors for OS in patients with myelofibrosis.

Momelotinib is a potent, selective, and orally-bioavailable JAK1, JAK2, and ACVR1 inhibitor with a differentiated therapeutic profile in myelofibrosis. As previously reported in the SIMPLIFY-1 (NCT01969838) and SIMPLIFY-2 (NCT02101268) trials, the drug induced robust OS in both patients who have received JAK inhibitors and those who are JAK inhibitor-naïve. Oh added that momelotinib demonstrated substantial anemia benefits compared with ruxolitinib including improved hemoglobin levels, sustained TI, and reduced need for transfusions.^2,3

“Importantly, momelotinib randomized patients who achieved or maintained transfusion independence at week 24, which we refer to as TI responders, had significantly improved survival compared to non TI responders,” he said.

Oh and his colleagues sought to examine the relationships between baseline hemoglobin, ferritin, hepcidin, and C-reactive protein, and week 24 response in the SIMPLIFY patient populations. Investigators employed logistic regression models to identify factors and/or thresholds that maximize the differential TI response across treatment arms in SIMPLIFY-1, then applied those thresholds to SIMPLIFY-2.

When investigators analyzed 370 SIMPLIFY-1 patients, they found that baseline ferritin was inversely related to baseline hemoglobin. Levels of each were comparable between treatment groups. “Ferritin was a strong predictor with a large dynamic range,” Oh said.

At baseline, the mean hemoglobin was 10.6 g/dL, and the median was 10.2 g/dL in the momelotinib group (n = 185) compared with 10.8 g/dL and 10.7 g/dL, respectively, in the ruxolitinib group (n = 185). Median ferritin was 136.1 ng/mL and mean was 384.6n ng/mL in the in the momelotinib group compared with 125.3 ng/mL and 341.0, respectively, with ruxolitinib.

The week 24 TI-R treatment effect of momelotinib (62%) vs ruxolitinib (35%) was significantly greater in patients with ferritin greater than or equal 90 ng/mL (response ratio [RR], 1.8; P = .0051) than in those with ferritin levels below 90ng/mL (79% and 73%, respectively, RR, 1.1).

Investigators observed this same trend in SIMPLIFY-2, where TI-R treatment effect of momelotinib vs best available therapy (BAT)/ruxolitinib was greater in the ≥90ng/mL cohort (41% vs 11%, RR, 3.8) than in the <90ng/mL cohort (57% and 50%, RR, 1.1), However, the interaction was marginally significant (P = .0707), possibly because of the smaller sample size.

Baseline hemoglobin levels also predicted week 24 TI-R favoring momelotinib at 9 g/dL to 12 g/dL (OR, 2.27; 95% CI, 0.77-0.66) and less than 9 g/dL (OR, 22.79; 95% CI, 1.23-421.88). Baseline ferritin levels also predicted week 24 TI-R at all hemoglobin levels less than or equal to 12 g/dL.

“Patients with the highest baseline ferritin had low week 24 TI responses with either treatment,” Oh said. “Similar trends were observed in SIMPLIFY-2 where the only week 24
TI responses in patients with the highest baseline ferritin levels occurred and those randomized to momelotinib.”

The SIMPLIFY trials comparing momelotinib versus ruxolitinib (Jakafi) or BAT included a total of 558 patients. Participants entered a 24-week randomization treatment phase, followed by an opportunity for extended momelotinib treatment for all patients. Patients in SIMPLIFY-1 were naïve to ruxolitinib while patients in the second trial had received prior treatment. Splenic response rate was the primary endpoint in both studies.

In results presented at the 2020 ASH Annual Meeting, in SIMPLIFY-1, the median OS was not reached for patients initially treated with momelotinib compared with 53.1 months in patients who crossed over from the ruxolitinib arm (HR = .99; P = .97). In results from SIMPLIFY-2 published in 2018, the median OS was 34.3 months for patients originally assigned to momelotinib and 37.5 months (HR = .96; P = .86) for those assigned to ruxolitinib/BAT who subsequently crossed over to the experimental drug. Lead author Srdan Verstovsek, MD, PhD, chief of the section for myeloproliferative neoplasms in the Department of Leukemia at The University of Texas MD Anderson Cancer Center, called that the best OS reported for patients previously treated with ruxolitinib.^2,3

The median OS was 37.5 months for those who crossed over to receive momelotinib.

References

1. Oh ST, Gerds A, Mesa R, et al. Baseline serum ferritin differentially predicts W24 transfusion independence response for momelotinib and ruxolitinib in patients with myelofibrosis. Presented at: 2021 ASH Annual Meeting and Exposition; December 11-14, 2021; Atlanta, GA. Abstract 3638.
2. Verstovsek S, Egyed M, Lech-Maranda E, Sacha T, et al. Robust overall survival and sustained efficacy outcomes during long term exposure to momelotinib in jak inhibitor naïve and previously JAK inhibitor treated intermediate/high risk myelofibrosis patients. Presented at: the 62nd ASH Annual Meeting; virtual. December 5, 2020.
3. Harrison CN, Vannucchi AM, Platzbecker U, et al. Momelotinib versus best available therapy in patients with myelofibrosis previously treated with ruxolitinib (SIMPLIFY 2): a randomised, open-label, phase 3 trial. Lancet Haematol. 2018;5(2):e73-e81. doi: 10.1016/S2352-3026(17)30237-5