Folate deficiency anemia in patients on PARP inhibition: Insights from a large single-center study.

Abstract

e24105 Background: Poly ADP-ribose polymerase inhibitors (PARPi) constitute a novel therapeutic class for BRCA1/2 -associated malignancies. Severe folate deficiency (FD) anemia has been reported in PARPi-treated patients (pts), leading to morbidity and mortality due to therapy disruptions and avoidable medical interventions. We conducted the first large retrospective study characterizing FD during PARPi use. Methods: This was a retrospective single-center analysis of PARPi-treated adults within the Mount Sinai Health System (1/1/18 - 10/10/23). De-identified clinical data was obtained from the Mount Sinai Data Warehouse and manual chart review. Fisher’s exact test was used for significance analysis. Summary statistics were generated using R v4.3.3. Results: 512 pts were treated with PARPi for ovarian (62.7%), breast (31.4%), fallopian tube (17.4%), prostate (14.3%), or pancreatic (4.3%) cancer. Of these, 16 (3.1%) experienced new FD to a nadir folate of < 2.2 - < 3.3 ng/mL within a median of 16.1 (IQR 8.0-38.2) weeks of olaparib (N = 15) or niraparib (N = 1) initiation. None had known malabsorptive risk and only one had a potential concurrent iatrogenic culprit. FD anemia caused PARPi disruptions in 12 (75%), including dose reduction (2 [12.5%]), hold (7 [43.8%]), and/or discontinuation (6 [37.5%]). Grade 4 anemia requiring red blood cell transfusions occurred in 62.5% (10/16) of those who developed FD, compared to 14.5% (72/496) of those who did not (p < 0.001). Among the folate deficient, none received folic acid (FA) at PARPi initiation. All nine who were re-tested achieved replete folate stores with oral supplementation, with successful PARPi resumption in five (71.4%) and re-escalation from reduced to full dose PARPi in one (50.0%). In another, earlier folate testing upon grade 2 anemia onset allowed for FA repletion with continued full-dose PARPi. An additional 152 (29.7%) experienced new macrocytic anemia on PARPi, unexplained by other common causes, but with unknown folate levels. Conclusions: This study is the first to systematically demonstrate the prevalence of FD anemia in association with PARPi use. In the United States population, FD is present in only < 0.1% of adults. Within our cohort of PARPi recipients, 3.1%—and potentially many more given another 29.7% with new unexplained macrocytic anemia—experienced clinically significant FD anemia on PARPi. Once diagnosed, supplementation effectively restored folate, corrected the associated anemia, and allowed for continued PARPi. Limitations of this study include its retrospective observational nature. This study highlights the importance of high clinical suspicion and early diagnosis of FD in PARPi-treated pts. To avoid this toxicity and its related transfusion needs and cancer treatment disruptions, regular folate testing and the low cost, low risk intervention of prophylactic FA supplementation should be considered in all PARPi recipients.