Milademetan in Advanced Solid Tumors with MDM2 Amplification and Wild-type TP53: Preclinical and Phase II Clinical Trial Results

Abstract

Purpose: Mouse double minute 2 (MDM2) is an E3 ubiquitin ligase that degrades the tumor suppressor p53. In cancers, MDM2 amplification (MDM2amp) leads to overexpression of MDM2, inducing p53 degradation and a p53-null phenotype even in the absence of TP53 mutations. We report here the preclinical and clinical activities of milademetan, a potent and selective oral small-molecule inhibitor of the MDM2-p53 interaction, in MDM2amp, TP53 wild-type (WT) solid tumors.

Patients and methods: Milademetan was tested against a variety of cell line and xenograft tumor models. This supported a phase II basket study (MANTRA-2) in patients with advanced MDM2amp, TP53-WT solid tumors. The primary endpoint was the objective response rate, and key secondary endpoints included progression-free survival and adverse events.

Results: Milademetan showed potent activity against MDM2amp, TP53-WT laboratory models. In the phase II trial, 40 patients received milademetan, 31 of whom had centrally confirmed molecular testing. The best overall response was 19.4% (6/31) with one confirmed response (3.2%) and five unconfirmed partial responses, including a patient with endometrial stromal sarcoma who achieved a 100% target lesion reduction. The median progression-free survival was 3.5 months (95% confidence interval, 1.8-3.7). Grade 3 or 4 adverse events observed included thrombocytopenia, neutropenia, anemia, leukopenia, and diarrhea.

Conclusion: Milademetan had a manageable safety profile and achieved responses against a variety of refractory MDM2amp, TP53-WT solid tumors, but tumor reductions were short-lived. Subsequent efforts should focus on combination strategies, further biomarker refinement, or novel MDM2 targeting approaches to achieve more durable clinical benefit.

Figure 1.

Activity of milademetan in cell lines. A, Milademetan had antiproliferative activity among cell lines with and without TP53 mutations (n = 3). B, Milademetan displayed potent activation of p53 target genes p21 and PUMA and apoptosis in a dose- and time-dependent manner. CN, copy number; LPS, liposarcoma.

Figure 2.

Activity of milademetan in PDX models. A, Milademetan induced antitumor activity in nude mice bearing LU-01-0448 PDX tumors using daily dosing of 100 mg/kg; ST-02-0075 gastric PDX models using daily dosing of 25, 50, and 100 mg/kg; LD1-0025-217643 lung adenocarcinoma PDX models using daily dosing of 100 mg/kg; and LD1-0025-217621 lung adenocarcinoma PDX models using daily dosing of 50 mg/kg. B, LU-01-0448 PDX tumor-bearing mice (n = 3) were treated orally with milademetan 100 mg/kg for 7 days, and tumor tissues were collected at indicated timepoints after the last dose. Tumor p21 and PUMA protein levels and plasma GDF-15/MIC-1 levels were evaluated. CN, copy number; LUAD, lung adenocarcinoma.

Figure 3.

Efficacy results on MANTRA-2. A, Waterfall plot of best tumor size response on MANTRA-2. Select co-mutations and MDM2 copy number on central testing are displayed; complete co-mutations are provided in Supplementary Fig. S6. B, Swimmer plot showing duration of therapy on study. cPR, confirmed partial response; GEJ, gastroesophageal junction; uPR, unconfirmed partial response.

Figure 4.

Association between global chromosomal instability and clinical benefit from milademetan. The fraction of the xT panel altered was defined as the fraction of genes on the central Tempus xT targeted sequencing assay with <1 or >3 copies. The association was statistically significant via the Kruskal–Wallis test, with P = 0.04. BOR, best overall response; NA, not amplified; PD, progressive disease; PR, partial response; SD, stable disease.