Therapy-related myeloid neoplasms (tMN) are a fatal consequence of exposure to oncologic therapy. Prior work has demonstrated that the expansion of TP53 clonal hematopoiesis (CH) under selective pressure of cytotoxic therapy is a major driver of tMN. Currently however, there are no strategies to reduce tMN risk other than avoidance of cytotoxic therapy. CDK4/6 are critical regulators of cell-cycle progression in hematopoietic stem and progenitor cells (HSPCs). Among functionally CDK4/6 independent tumors (such as those with loss of Retinoblastoma, representing ~10% of all cancers), CDK4/6 pharmacologic inhibition has been shown to induce quiescence in normal HSPCs and reduce chemotherapy-induced bone marrow suppression without interfering with the anti-tumor efficacy of chemotherapy. Trilaciclib is an intravenously administered, CDK4/6 inhibitor that is FDA-approved to decrease the incidence of chemotherapy-induced myelosuppression in patients with small cell lung cancer (SCLC). We hypothesized that by protecting HSPCs from the cytotoxic effects of chemotherapy, trilaciclib might also reduce the clonal expansion of TP53-mutant CH during chemotherapy.

We obtained serial blood samples from healthy controls (n=176) and three placebo-controlled randomized clinical trials of trilaciclib including patients with (1) SCLC (n=65) receiving carboplatin and etoposide, (2) metastatic colorectal cancer (mCRC) (n=125) receiving leucovorin/fluorouracil/oxaliplatin/irinotecan (FOLFOXIRI) plus bevacizumab, and (3) metastatic triple negative breast cancer (mTNBC) (n=34) receiving gemcitabine and carboplatin. We sequenced peripheral blood genomic DNA at treatment onset and after several rounds of chemotherapy using deep targeted UMI-based sequencing. Across all three clinical trials, DNA damage response (DDR) CH clones, including TP53, expanded more rapidly with chemotherapy in both trilaciclib groups (SCLC p=7.2e-5, mCRC p=2.7e-3, mTNBC p=1.4e-3) and placebo groups (SCLC p=2.2e-13, mCRC p=1.2e-8, mTNBC p=1.3e-6) compared to untreated control individuals. However, CH growth rate was significantly lower in the trilaciclib arm compared to the placebo (SCLC p=9.5e-3, mCRC p=1.8e-5, mTNBC p=0.045).

To evaluate the functional effect of CDK4/6 inhibition, 8-week-old CD45.1+ recipient C57/BL6 mice were lethally irradiated and engrafted with bone marrow (BM) cells composed of a 9:1 ratio of CD45.1+ Vav-cre p53 wild-type (WT) cells and CD45.2+ Vav-cre p53R172H/WT cells or CD45.2+ Vav-cre p53 WT cells. After a month, mice were randomized to receive vehicle, carboplatin (30 mg/kg), trilaciclib (100 mg/kg), or carboplatin with trilaciclib (given 30 min prior) for 4 weeks. Carboplatin treatment expanded the p53R172H/WT cell population from a mean of 18% to 45% in the peripheral blood with similar expansion of bone marrow p53 mutant long-term hematopoietic stem cells (LT-HSCs). However, simultaneous administration of trilaciclib and carboplatin abrogated this growth. Evaluation of peripheral blood and marrow chimerism 6 weeks after cessation of drug treatment revealed that the suppression of p53 mutant CH expansion with trilaciclib use before carboplatin persisted after treatment. Evaluation of apoptosis in chimeric mice revealed that concomitant treatment with CDK4/6 inhibition and carboplatin increased apoptosis in p53 mutant relative to WT LT-HSCs. These data suggest a selective effect of CDK4/6 inhibition in promoting the death of p53 mutant cells and the preservation of WT HSPCs. Similar effects were seen with the oral CDK4/6 inhibitor palbociclib and using a dominant negative form of CDK6 introduced into p53 mutant HSPCs.

Single cell RNA-seq of chimeric mice treated with carboplatin with trilaciclib or single agent controls revealed that CDK4/6 inhibition treatment promotes HSC and myeloid progenitor quiescence while mitigating the myeloid cell expansion and increased stemness advantage of Trp53 mutant HSPCs observed in the setting of cytotoxic chemotherapy.

Overall, these data demonstrate that CDK4/6 inhibition given in conjunction with a variety of chemotherapy regimens and across diverse cancer populations mitigates chemotherapy-related expansion of CH clones with mutations in DNA damage response genes. This finding was also observed in a syngeneic murine model of TP53 mutant CH. This represents the first demonstration in patients of a pharmacologic strategy to block chemotherapy-induced expansion of pre-leukemic TP53-mutant clones.

This content is only available as a PDF.
Sign in via your Institution