Intermittent Dosing Schedule of MDM2 Inhibitor Appears Tolerable in Advanced Cancers

The selective small molecule MDM2 inhibitor milademetan (RAIN-32) administered at an intermittent dosing schedule was tolerable and showed early efficacy in patients with advanced cancers, according to findings from a phase 1 study (NCT01877382).¹

Common grade 3/4 milademetan-related adverse effects (AEs) included thrombocytopenia (29.0%), neutropenia (15.0%), and anemia (13.1%). The rates of these grade 3/4 AEs in patients receiving the recommended intermittent dose and schedule of milademetan were 15.0% (n = 3), 5.0%, and 0%, respectively.

“An intermittent dosing schedule of 3/14 days of milademetan mitigates dose-limiting hematologic abnormalities while maintaining efficacy,” lead study author, Mrinal M. Gounder, MD, of Memorial Sloan Kettering Cancer Center in New York, New York, and colleagues, wrote.

Tumors such as dedifferentiated liposarcomas and intimal sarcomas have MDM2 gene amplification. In preclinical studies, MDM2 inhibition with milademetan led to p53-induced apoptosis in cancer cell lines.² This may restore p53 tumor suppressor activity.

This 2-part, open-label, first-in-human study aimed to investigate the safety, pharmacodynamics, pharmacokinetics, and preliminary efficacy of milademetan at various dosing schedules in patients with advanced solid tumors or lymphomas.¹ The primary end points of part 1, dose-escalation (n = 87), were the safety and tolerability of milademetan to determine the maximum tolerated dose or recommended phase 2 dose. Secondary end points in part 1 were tumor response, pharmacokinetics, and pharmacodynamic effects on GDF15. Efficacy end points included overall response rate (ORR), time to response, duration of response, disease control rate (DCR), and progression-free survival (PFS).

The primary end point of part 2, dose-expansion (n = 20), was confirmation of the safety and tolerability of milademetan in patients with diffuse large B-cell lymphoma (DLBCL) or advanced melanoma. However, part 2 was closed upon determining that it could not meet its objectives with the dosing schedule used. Instead, the investigators decided to evaluate alternative dosing schedules in an expanded dose-escalation phase in patients with dedifferentiated liposarcomas.

Eligible patients included those at least 18 years of age with advanced solid tumors or lymphomas. Patients with tumor types with a high prevalence of MDM2 overexpression or amplification were preferentially enrolled in part 1, although tests to determine MDM2 amplification status were not performed. Patients needed to have an ECOG performance status of 0 or 1, adequate bone marrow function, a hemoglobin level at or above 9.0 g/dL, an absolute neutrophil count of at least 1.5 × 109/L, blood clotting, and renal and hepatic function.

Patients with known TP53 mutations were excluded. However, patients who were subsequently confirmed to have TP53 mutations after initiating milademetan dosing were allowed to remain on the study if they were deriving clinical benefit.

Patients received oral milademetan once daily at a starting dose of 15 mg in 28-day cycles on extended/continuous schedules (days 1-21 or days 1-28) or intermittent schedules (days 1-7 or days 1-3 and 15-17) when the extended/continuous schedules led to excessive toxicity. Treatment was continued until disease progression, withdrawal of consent, or unacceptable toxicity. In patients who experienced AEs but derived clinical benefit, dose reductions by 1 dose level were allowed once the toxicity lessened to grade 1 or lower. Treatment was discontinued in patients who required over 4 weeks to recover from acute toxicities and in patients who derived clinical benefit but required over 8 weeks to recover from acute toxicities.

Between July 2013 and August 2018, 107 patients were enrolled. Four patients withdrew consent before beginning treatment. At primary analysis, all patients had discontinued study treatment, 69.2% (n = 74) because of disease progression. Additionally, 4.7% of patients (n = 5) who had not progressed by study close continued milademetan treatment through a post-trial access program. The median duration of treatment was 2.6 months (range, 0.1-50.9).

The most common cancers were dedifferentiated liposarcomas (49.5%; n = 53), melanoma (20.6%; n = 22), and lymphoma (3.7%; n = 4). Overall, 61.7% of patients (n = 66) had received at least 3 prior systemic therapies. Additionally, 66.4% of patients (n = 71) had TP53 wild-type tumors, and 12.1% (n = 13) had confirmed TP53-mutant tumors.

No dose-limiting toxicities (DLTs) occurred with milademetan at 15 mg, 30 mg, or 60 mg. Hematologic DLTs occurred at 120 mg (grade 4 thrombocytopenia, n = 1), 160 mg (grade 4 grade 4 neutropenia, grade 2 thrombocytopenia, and grade 2 leukopenia, n = 1), and 240 mg (grade 4 thrombocytopenia, n = 1).

At the intermittent dosing schedule, DLTs occurred at 120 mg once daily on day 7 of 28 (grade 2 malaise and fatigue, n = 1) and 340 mg once daily on days 1 through 3 and 15 through 17 (grade 3 thrombocytopenia, n = 1). The investigators determined 260 mg once daily on the intermittent schedule, days 1 through 3 and 15 through 17 of each 28-day cycle, to be the recommended dose for future development of milademetan.

The most common all-grade milademetan-related AEs were nausea (72.0%), thrombocytopenia (60.7%), fatigue (44.9%), and anemia (35.5%). In patients who received milademetan on the intermittent schedule (n = 38), the respective rates of all-grade and grade 3/4 thrombocytopenia were 44.7% and 15.8%. In those who received treatment on the extended/continuous schedules (n = 69), the respective rates of all-grade and grade 3/4 thrombocytopenia were 69.6% and 36.2%.

Of the patients treated on the intermittent schedule, 21.1% (n = 8) required dose reductions and 15.8% (n = 6) required dose interruptions for drug-related thrombocytopenia. Of the patients treated on the extended/continuous schedule, 23.2% (n = 16) required dose reductions and 34.8% (n = 24) required dose interruptions for drug-related thrombocytopenia. Additionally, 11.6% of patients (n = 8) treated on the extended/continuous schedule experienced drug-related serious AEs.

Of the patients who received milademetan at the recommended dose and schedule, 40.0% (n = 8) required dose reductions for milademetan-related AEs, including thrombocytopenia (n = 7) and vomiting (n = 1). Additionally, 25.0% (n = 5) of these patients required dose interruptions because of drug-related AEs, including thrombocytopenia (n = 2), thrombocytopenia and conjunctivitis (n = 1), thrombocytopenia and nausea (n = 1), and neutropenia (n = 1). These dose interruptions had a median duration of 21 days per episode (range, 5-62).

At the recommended dose, the median time to maximum serum concentration was 3.1 hours. On day 1 of cycle 1, the geometric mean maximum serum concentration was 1503 ng/mL and the area under the curve0-24 was 18,432 ng*h/mL. The geometric mean apparent total clearance was 15.6 L/h and the terminal elimination half-life was 10.0 hours.

Across all schedules, milademetan increased serum GDF15 levels with increasing plasma concentrations, with a mean increase of up to 10-fold over baseline within 24 hours after the first dose. By day 8 of cycle 1, milademetan had increased p53, p21, and MDM2 expression levels in the 6 evaluable samples available at that time.

Two patients each with dedifferentiated liposarcomas and non-liposarcoma tumors achieved partial responses (PRs). The ORR was 3.8% (n = 2; 95% CI, 0.5%-13.0%) in patients with dedifferentiated liposarcomas (n = 53) and 5.9% (n = 2; 95% CI, 0.7%-19.7%) in those with non-liposarcoma tumors (n = 34).

Across all cohorts, the DCR was 45.8% (95% CI, 36.1%-55.7%). In patients with dedifferentiated liposarcomas (n = 53), the DCR was 58.5% (95% CI, 44.1%-71.9%) in the overall population and 62.0% (95% CI, 35.4%-84.8%) in those who received the recommended intermittent schedule (n = 16). In patients with non-liposarcoma tumors, the DCR was 32.4% (95% CI, 17.4%-50.5%).

In all patients, the median PFS was 4.0 months (95% CI, 3.4-5.7). In patients with dedifferentiated liposarcomas, the median PFS was 7.2 months (95% CI, 3.8-10.1) in the overall population, 6.3 months (95% CI, 3.8-10.0) in those who received the extended/continuous schedules (n = 30), and 7.4 months (95% CI, 2.7-14.6) in those who received the intermittent schedule (n = 23). In those with dedifferentiated liposarcomas who received the recommended dose and schedule, the median PFS was 7.4 months (95% CI, 1.8-14.6). In those with previously treated dedifferentiated liposarcomas (n = 11), the median PFS was 8.0 months (95% CI, 1.8-27.7). In patients with non-liposarcoma tumors, the median PFS was 3.4 months (95% CI, 1.8-5.6).

In an exploratory analysis to determine median PFS by line of therapy for all patients with dedifferentiated liposarcomas, the median PFS was 14.6 months (95% CI, 3.8-not estimable) in previously untreated patients (n = 17) and 5.9 months (95% CI, 3.5-10.0) in patients with at least 1 prior therapy (n = 36).

In the part 2 dose-expansion cohort, 5.6% (n = 1) of patients with melanoma (n = 18) achieved a PR and 44.4% (n = 8) achieved stable disease. Both patients with DLBCL were unevaluable. The ORR for the total cohort was 5.0% (95% CI, 0.1%-24.9%). The DCR was 35.0% (95% CI, 15.4%-59.2%) and the median PFS was 3.2 months (95% CI, 1.6-4.0).

“On the basis of our observations, a randomized, phase 3 registration study (MANTRA; RAIN-3201) [NCT04979442] of milademetan vs trabectedin in patients with unresectable or metastatic dedifferentiated liposarcomas with disease progression on at least 1 prior systemic therapy has recently started enrolling patients,” the study authors concluded.

References

1. Gounder MM, Bauer TM, Schwartz GK, et al. A first-in-human phase I study of milademetan, an MDM2 inhibitor, in patients with advanced liposarcoma, solid tumors, or lymphomas. J Clin Oncol. Published online January 20, 2023. doi:10.1200/JCO.22.01285
2. Ishizawa J, Nakamaru K, Seki T, et al. Predictive gene signatures determine tumor sensitivity to MDM2 inhibition. Cancer Res. 2018;78(10):2721-2731. doi:10.1158/0008-5472.CAN-17-0949