Mitigating Therapy-Related Myeloid Neoplasia in p53-Mutant Clones through Targeted DNA Repair Pathway Inhibition with PARP Inhibitors

Therapy-related myeloid neoplasms (t-MNs), especially therapy-related acute myeloid leukemias (t-AML), pose a severe risk among cancer survivors, notably those treated for ovarian and breast cancers. Given the inherent resistance of t-MNs to conventional cytotoxic treatments due to their induction by chemotherapy, there is a critical need for innovative treatment approaches. Targeting DNA repair pathways, especially with PARP inhibitors (PARPi), offers a promising approach. PARPi, by exploiting the DNA repair deficiencies of cells with defective p53, can selectively inhibit the growth and survival of these malignant clones. However, there is growing evidence linking PARPi, often used in combination with chemotherapy, to the development of t-MNs. This association raises concerns regarding the sequence and timing of these therapies in cancer treatment protocols. However, the implications of administering PARPi alone before the initiation of chemotherapy, have not been thoroughly investigated.

To evaluate the effect of PARPi on p53-mutant clones, we established a mouse model of p53-mutated clonal hematopoiesis. We utilized a chimeric bone marrow (BM) transplantation method, where BM cells that carry the Trp53^R172H mutation, isolated from Vav-Cre;Trp53^wmR172H/fl;mTmG mice and identifiable by their GFP expression, were mixed with congenic wild-type BM cells from mTmG mice, marked by RFP, and transplanted into irradiated hosts. This system, using dual markers, enables effective tracking and differentiation between the mutant and wild-type cell populations. Six weeks after transplant, the levels of p53-mutant cells in the peripheral blood of recipient mice were measured using flow cytometry. Mice were divided into four groups, each containing 10 mice: one control group received no treatment, while the other groups were treated with Olaparib, carboplatin, or a combination of both for three weeks.

Analysis of peripheral blood 2 months after treatment revealed a significant decrease in the p53-mutant cell population in the Olaparib group compared to the control group. In contrast, both the carboplatin group and the combination therapy group showed a significant increase in these clones, which eventually led to the development of t-AML with full penetrance. CyTOF analysis of three mice in each group further confirmed the efficient targeting of p53-mutant progenitor cells. BM analysis three months after treatment initiation revealed the absence or near-absence of cells expressing mutant p53 in Olaparib group, whereas in the carboplatin and the combination therapy group, the expanding clones completely occupied the BM. None of the mice in the Olaparib-only group developed leukemia, but those in the combination therapy group succumbed to leukemia, similar to those treated with carboplatin alone. Western blot analysis revealed the upregulation of pH2AX, p53 and Cyclin D2 in the carboplatin group. scRNA-seq analysis showed that the Olaparib-treated group exhibited downregulation of genes associated with DNA damage response compared to the control group. Together, the data indicate that Olaparib effectively targets and reduces or eliminates p53-mutant clones, potentially preventing the onset of leukemia, whereas combination therapy does not prevent leukemia progression and mirrors the outcomes seen with carboplatin alone.

Our findings establish PARP inhibition as a novel therapeutic strategy for managing p53 mutant clonal hematopoiesis to prevent t-MNs. Specifically, PARP inhibition with Olaparib can effectively reduce the population of p53-mutant clones and prevent the onset of leukemia. However, combining Olaparib with carboplatin exacerbates clonal expansion and lead to t-AML, underscoring the importance of the sequence and timing of these therapies. Our data indicate that PARP inhibition alone can delay or prevent the progression from p53 mutant clonal hematopoiesis to AML.