Leukemia: New Adverse Event With PARP Inhibitors in Cancer

Pam Harrison

January 07, 2021

A new adverse event has been reported for PARP inhibitor drugs used in the treatment of cancer: the risk of developing myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML), both of which have a high mortality rate.  

PARP inhibitors are increasingly being used for cancer, particularly in patients with ovarian cancer, as well as some breast cancer and prostate cancer in patients with germline BRCA mutations.

A new analysis shows that risk of developing MDS or AML is more than 2.5 times higher in patients on a PARP inhibitor when compared with placebo.

The finding comes from a meta-analysis of clinical trials involving a total of 9099 patients, many of whom had ovarian cancer, and the drugs included olaparib (Lynparza) as the most frequently used, followed by veliparib (ABT-888), niraparib (Zejula), rucaparib (Rubraca) and talazoparib (Talzenna).

The risk was also uncovered in a companion analysis of the World Health Organization's real-world pharmacovigilance database, VigiBase.

Both are detailed in a paper published online December 18 in Lancet Haematology.

"To our knowledge, this large-scale analysis is the first to have shown an increased risk of myelodysplastic syndrome and acute myeloid leukemia related to PARP inhibitor treatment," write the authors, led by Joachim Alexandre, MD, PhD, Caen University Hospital, Caen, France.

"Clinicians should be aware of these potentially fatal hematological toxicities, particularly in the front-line maintenance setting," they add.

The new findings are unlikely to change the clinical management of ovarian cancer given that the risk of MDS/AML is still relatively low, comments a specialist in gynecologic cancers, Anna Tinker, MD, University of British Columbia, Vancouver, Canada, in an accompanying editorial.  

But she emphasizes that "clinicians need to remain aware that the onset of myelodysplastic syndrome and acute myeloid leukemia can occur soon after PARP inhibitor initiation even after brief drug exposure."

"They are definitely a worry," Tinker told Medscape Medical News. "But women with advanced stage ovarian cancer face a 70% to 90% chance of having a future recurrence of their cancer which is a certain fatality," she noted. "So on balance, I think knowing there is a small risk of MDS or AML probably wouldn't change their decision to have a PARP inhibitor."

However, Tinker emphasized the importance of monitoring patients on a PARP inhibitor during therapy and following discontinuation of treatment, as these so-called "late toxicities" can theoretically happen at any point following initiation of treatment.

"As I pointed out in the editorial, MDS and AML are not necessarily a late toxicity. They can occur soon after initiation of a PARP inhibitor — in some cases within a few months — so the key thing is not to mislead patients by say this is a 'late' toxicity; it can arise at any time," she emphasized.

Thus, if a cytopenia does not resolve following withdrawal of treatment, "this should be your first sign that something is not quite right and further evaluation must be done," Tinker emphasized.

"The imperative remains to properly counsel patients on the rare but life-threatening toxicities of PARP inhibitors, and to offer appropriate monitoring and investigations of hematological changes," Tinker writes in the editorial.

However, she also emphasized that PARP inhibitors really are the best treatment option for patients with advanced ovarian cancer. Knowing about these adverse effects would not deter her from using these agents in this patient population, so long as there was appropriate vigilance throughout therapy, she noted.

Safety Meta-Analysis

The meta-analysis included 28 randomized controlled trials (RCTs) with available adverse events (AEs); 18 of the trials were placebo-controlled and 10 had other control arms. Of the 9099 trial participants, 63% received a PARP inhibitor, 37% were in the control groups. Importantly, 43% of the studies analyzed were carried out in patients with ovarian cancer.

Based on the 18 placebo-controlled RCTs involving over 7300 patients, PARP inhibitors were shown to increase the risk of myelodysplastic syndrome and acute myeloid leukemia by over 2.5-fold (odds ratio [OR], 2.63, P = .026), the investigators report.

Investigators noted that no heterogeneity was observed across studies. Of equal importance, they emphasize that "all myelodysplastic syndrome and acute myeloid leukemia cases were reported in RCTs in ovarian cancers."

The median latency period was 20.3 months, although this ranged from 18.4 to 26.6 months, they add.

Across all RCTs, including both placebo and non-placebo RCTs, the incidence of myelodysplastic syndrome and acute myeloid leukemia was 0.83% (P = 0.84). PARP inhibitor therapy also increased the risk of myelodysplastic syndrome and acute myeloid leukemia vs all control treatments by over twofold (OR, 2.25), with again no heterogeneity across studies, the authors observe.

"Subgroup analyses did not show significant differences with regard to previous systemic therapy, [type of] PARP inhibitor used…or PARP inhibitor treatment duration," investigators note.

VigiBase Analysis

Additional evidence comes from an analysis of data from the WHO's VigiBase. A total of 178 cases of MDS and AML related to PARP inhibitor use were identified in the database as of May 3, 2020.

"All cases were considered serious," the authors observe.

Some 85% of patients in the VigiBase analysis had been treated with a PARP inhibitor for ovarian cancer.

In a limited number of these cases, the median treatment duration was only 9.8 months, ranging from 6 days to 5½ years. The median latency period between the emergence of myelodysplastic syndrome and acute myeloid leukemia from first exposure to a PARP inhibitor was 17.8 months, as investigators note, although the median duration from first PARP inhibitor exposure to the emergence of acute myeloid leukemia was slightly longer at 20.6 months, they add.

Outcomes were not available for all the cases identified in the VigiBase analysis, but of those that were reported, 45% resulted in death.

Overall, cytopenia was reported in 40% of patients who developed MDS or AML, the most frequent type being anemia. Patients experienced cytopenia at a median of 7 months after initiation of PARP inhibitor therapy.

Given these real-world results from the pharmacovigilance cohort, the authors recommend that if patients have not recovered within 28 days or have persistent cytopenia following dose modification, further investigation including bone marrow analysis and blood analysis for cytogenetics must be done. Patients must also be monitored monthly and consideration should be given to discontinuing PARP inhibitor therapy.

Most importantly, "PARP inhibitor [therapy] must be discontinued if myelodysplastic syndrome or acute myeloid leukemia, or secondary cancers, are confirmed," they stress.

Patients With and Without BRCA Mutations

Interestingly, analysis of data from the placebo-controlled RCTs showed that patients who did not have either a BRCA1 or BRCA2 mutation, or homologous recombination deficiencies, clinically benefited from PARP inhibitor maintenance treatment, even if this benefit was lower than it was in patients with either BRCA mutation or homologous recombination deficiencies.

"These data highlight [the fact] that patients without clearly identified biomarkers could be eligible for PARP inhibitors therapies and require a case-by-case risk–benefit assessment," the authors suggest.

"And clinicians need to remain vigilant in evaluating delayed hematological toxicities in patients treated with PARP inhibitors," they conclude.

Alexandre has disclosed no relevant financial relationships, but many of the study coauthors have financial ties to industry, as listed in the original article.

Tinker reports receiving grants, honoraria and speaker fees from AstraZeneca and speaker fees from GlaxoSmithKline.

Lancet Haematol. Published online December 18, 2020. Abstract, Editorial

For more from Medscape Oncology, join us on Twitter and Facebook


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.