PARP Inhibitors and Secondary Leukemia in Ovarian Cancer

New research in women with ovarian cancer confirmed an increased risk of myelodysplastic syndromes (MDS) or acute myeloid leukemia (AML) after poly(ADP-ribose) polymerase (PARP) inhibitor treatment and identified potential factors that may help predict individual risk.

How Large Is the Secondary Blood Cancer Risk with PARP Inhibitors?

At the Society of Gynecologic Oncology 2024 Annual Meeting on Women's Cancer (the society's 2024 annual meeting on women's cancer), Dr. Amma Asare of the University of Texas MD Anderson Cancer Center suggested that PARP inhibitor-associated myeloid neoplasms may have "unique clinical or molecular associations" that could help identify risk.

Individual risk matters because secondary hematologic malignancies associated with PARP inhibitors have particularly poor outcomes. Predictive ability currently "remains limited."

The team identified 32,356 patients treated at their institution for any gynecologic cancer between 2000 and 2022.

The secondary MDS/AML rate in the entire population was approximately 1% , consistent with earlier reports, Asare noted. Among ovarian cancer patients receiving a PARP inhibitor, however, it was approximately ten times higher (9%, or 32 cases among 355 patients).

Compared with patients without secondary leukemia, those who developed it had a longer interval between ovarian cancer diagnosis and starting a PARP inhibitor. There was no association with total treatment days or age at initiation.

Regarding chemotherapy, fewer PARP inhibitor-treated patients who developed secondary leukemia had received neoadjuvant chemotherapy.

Those with secondary leukemia had also received more carboplatin cycles and more chemotherapy lines overall.

Looking at patient-specific factors, Asare and colleagues found that those developing secondary leukemia had, compared with all other mutation types, more germline BRCA1 mutations represented in the group.

Patients developing secondary leukemia also had a lower platelet nadir during PARP inhibitor treatment. There was no association with pretreatment platelet nadir or inhibitor type.

Multivariable analysis identified three important factors:

1. Neoadjuvant chemotherapy was associated with a protective effect (OR, 0.24; P = 0.017),
2. Low platelets during PARP inhibitor treatment (OR, 0.99; P = 0.014) and
3. More total carboplatin cycles (OR, 1.15; P = 0.13) were associated with secondary leukemia.

Molecular analysis found TP53 mutations in 68% of PARP inhibitor-treated patients who developed a secondary hematologic cancer.

These data support roles for both genetic and environmental factors in PARP [inhibitor]-associated hematologic malignancy.

Platelet declines during treatment may be "a potential indicator of future MDS/AML risk." They may inform individual risk prediction and evidence-based management and monitoring.

Below is a mind map of ovarian cancer and PARP inhibitors:

Balancing Risks and Benefits

Secondary leukemia risk did not differ by PARP inhibitor type.

The absence of an association between treatment duration and MDS/AML risk differs from some other trials.

TP53 enrichment in those developing leukemia is also "quite interesting," raising the question of whether TP53 clonal hematopoiesis could mark increased MDS and AML risk with PARP inhibition.

Overall, available data indicate that the potential risks and toxicities of PARP inhibitors require careful consideration. It is essential to balance their benefits against the risks they may bring. In addition, to make the best use of these treatments and identify their benefits more precisely, more advanced biomarkers are needed.