As-Expected Ruxolitinib Starting Dose Outperforms Lower Dosage in Myelofibrosis

The arm receiving the expected dose of ruxolitinib did not reach a median OS.

Patients with myelofibrosis who were treated with the expected starting dose of ruxolitinib (Jakafi) demonstrated higher spleen response rates and a shorter median time to first response vs patients receiving a less-than-expected starting dose, per data from the ROMEI trial.

Additionally, overall survival (OS) was not reached in the as-expected group; however, patients in both groups reduced daily dosages over 12 months in the multicenter, observational, ongoing study, according to findings published in Cancer.

“To ensure optimal treatment with ruxolitinib, patients should be started on appropriate ruxolitinib doses and maintain the highest tolerated dose for maximum effectiveness,” wrote the study authors. “Patients who received the recommended doses showed a better trend in response.”

Trial Design and Key Data

The trial enrolled 508 adult ruxolitinib-naive patients who had been diagnosed with primary or secondary myelofibrosis across 51 centers in Italy between April 2017 and May 2022, at which time ruxolitinib was the only FDA-approved JAK inhibitor for myelofibrosis.

Patients with available baseline platelet counts were stratified into the as-expected (n = 174) and lower-than-expected (n = 132) dosage groups, resulting in 43% of patients receiving a lower-than-expected dose.

Changes in symptoms and health-related quality of life (HRQOL), assessed using Myeloproliferative Neoplasm 10 total symptom score (MPN-10 TSS) and EuroQol5-Dimension 5-Level (EQ-5D-5L), respectively, were primary end points of the study.

Response Rates and QOL

Spleen response was observed in 50.0% (95% CI, 39.0%-61.0%) and 30.2% (95% CI, 19.2%-43.0%) of patients in the as-expected and less-than-expected dose groups, respectively, at 24 weeks, and in 57.7% (95% CI, 46.0%-68.8%) and 45.8% (95% CI, 31.4%-60.8%) of patients at 48 weeks.

The as-expected dose group had a shorter median time to first spleen response with 3.3 months (95% CI, 1.9-6.6), vs 11.1 months (95% CI, 7.8-19.3; P = .019) in the less-than-expected dose group.

At 24 and 48 weeks, both patient groups’ MPN-10-TSS improved compared with baseline. The as-expected group had a greater proportion of patients with increased MPN-10-TSS vs the less-than-expected group from week 4 (38.5% vs 27.3%, respectively) to week 24 (40.8% vs 39.4%, respectively).

However, this changed by week 48, with 40.8% and 40.9% of patients in the as-expected and less-than-expected arms, respectively, reporting improved total symptom scores.

Sensitivity analysis without missing imputation revealed a greater proportion of patients with symptom response in the as-expected arm (50.4%) than the less-than-expected arm (43.6%) at week 48. At week 24, proportions were similar (44.8% vs 44.6%, respectively).

Both trial arms saw slight increases in HRQOL vs baseline at weeks 24 and 48. At 24 weeks, mean EQ-VAS scores for the as-expected group was 67.7 ± 19.4, compared to 62.9 ± 20.1 at baseline; the mean score increased to 65.3 ± 21.9 at 48 weeks. Likewise, in the less-than-expected dose group, the mean EQ-VAS score was 59.5 ± 18.9 at baseline, 67.1 ± 16.1 at 24 weeks, and 64.4 ± 19.1 at 48 weeks.

AEs, Interruption, and Discontinuation

Adverse events (AEs) were reported in 152 patients (87.4%) in the as-expected arm and 112 patients (84.9%) in the less-than-expected arm at data cutoff.

Rates of AEs were slightly higher in the as-expected dose group vs the less-than-expected dose group overall (87.4% vs 84.9%, respectively), and for the most commonly reported AEs, including blood and lymphatic disorders (63.8% vs 51.5%), anemia (50.6% vs 40.9%, and thrombocytopenia (29.9% vs 25.8%). Gastrointestinal disorders were reported at a lower rate in the as-expected group than the less-than-expected group (17.2% vs 24.2%).

The most common starting dose (43.7%) was 20 mg twice daily, followed by 22.3% receiving 10 mg twice daily, 20.6% receiving 15 mg twice daily, and 12.0% receiving 5 mg twice daily; 1.4% received alternative starting doses, including 5 mg daily, 11.3 mg daily, and 15 mg daily.

Similar rates of patients with at least 1 treatment interruption were reported in the as-expected and less-than-expected groups (10.3% vs 10.6%, respectively). The as-expected group had a shorter median time to first interruption (76.0 days) compared with the less-than-expected group (116.5 days).

Median treatment duration was higher in the as-expected group (51.2 months; 95% CI, 5.4-14.1) vs the less-than-expected group (33.3 months, 95% CI, 25.2-45.2). Permanent discontinuation rates at 6 and 12 months were lower in the as-expected group (8.7% [95% CI, 5.4%-14.1%] and 17.3% [95% CI, 12.3%-23.9%], respectively) than in the less-than-expected dose group (15.6% [95% CI, 10.3%-23.1%] and 27.6% [95% CI, 20.7%-36.3%]).

“The ROMEI study demonstrated the importance of optimal ruxolitinib dosage in patients with [myelofibrosis] for maximum effectiveness and improved OS, with manageable safety,” wrote investigators.

Five patients in the as-expected arm and 8 patients in the less-than-expected arm died during the trial.

Baseline Patient Characteristics

Myelofibrosis diagnoses were primary in 51.2% and 62.1% of the as-expected and less-than-expected arms, respectively. Additionally, most patients started treatment with ruxolitinib within 12 months of diagnosis, including 62.1% of those with as-expected dosages and 56.1% of those with less-than-expected dosages.

Patient groups also had similar proportions of patients with spleen lengths of at least 10 cm, with 35.2% in the as-expected group and 37.7% in the less-than-expected group (P = .672).

In the less-than-expected group, hemoglobin counts were significantly lower than in the as-expected group (10.8 g/dL vs 11.4 g/Dl, respectively; P = .009), as were platelet counts (290 x 10⁹/L vs 354 x 10⁹/L, respectively; P = .021).

Reference

Breccia M, Palandri F, Martelli M, et al. Dosing and clinical outcomes of ruxolitinib in patients with myelofibrosis in a real-world setting: Interim results of the Italian observational study (ROMEI). Cancer. 2025;131(7):e35801. doi:10.1002/cncr.35801