Analysis of Clinical Factors in Olaparib-related Anemia Using Adverse Drug Event Reporting Databases

Background/Aim: Anemia is one of the dose-limiting toxicities of olaparib. A global randomized controlled trial confirmed that anemia occurrence in Japanese was relatively high. The factors related to anemia in different nationalities remain unknown. Therefore, this study investigated the factors of olaparib-related anemia in real-world settings using an adverse event reporting system database. Patients and Methods: We used data from FDA Adverse Events Reporting System (FAERS) and Japanese Adverse Drug Event Report database (JADER) between 2018 and 2021. FAERS reports from Japan were collected to conduct subgroup analysis, which was defined as FAERS-Japan. The endpoint was the occurrence of olaparib-related anemia. Disproportionality analysis was conducted to calculate reporting odds ratio (ROR), with a confidence interval of 95%. Adjusted ROR (aROR) was calculated to control for sex differences. Results: In FAERS and JADER, the daily olaparib dose per body weight (DPBW) ≥12 mg/kg was associated with anemia occurrence [aROR; FAERS, 4.483 (3.009-6.680), p<0.001, FAERS-Japan, 1.834 (1.091-3.063), p=0.009, and JADER, 1.628 (1.039-2.551), p=0.034]. Furthermore, FAERS reports confirmed that females with body weight <50 kg, reports from Japan, concomitant use of drugs causing vitamin B12 deficiency, and previous platinum treatment history were associated with olaparib-related anemia. FAERS-Japan also showed that body weight <50 kg and previous platinum treatment history were associated with anemia occurrence. Conclusion: High DPBW constitutes a significant risk of olaparib-related anemia. In addition, information on co-administration of drugs causing vitamin B12 deficiency and previous platinum treatment history is also important for the evaluation of the risk of olaparib-related anemia.

administrative information, drug and biologic information, adverse events, patient outcomes, report sources, start end dates of drug therapy, and indications for use/diagnosis.JADER consists of four data tables: patient demographic information, drug information, adverse events, and primary disease.In JADER, age and body weight are categorized in 10-year and 10-kg increments, respectively; however, in our study, we used in the analysis the median value of each group.The source of the databases complies with the guidelines set by the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) and the databases adhere to ICH-standardized adverse event information guidelines (6).For this analysis, adverse event reports were downloaded from FDA websites (https://www.fda.gov/; last accessed on May 4 th , 2022) and the Pharmaceuticals and Medical Devices Agency (https://www.pmda.go.jp/; last accessed on May 4 th , 2022).Furthermore, FAERS reports reported from Japan and were defined as FAERS-Japan (7).Reports with the same case number were identified as duplicate reports and we used the most recent report as recommended by the FDA (8).
Collected data included case id, drug name, adverse event name, the start date of administration, the end date of administration, date of occurrence of adverse event, sex, age, body weight, reporting country and daily olaparib dose, while the daily olaparib dose per body weight (DPBW) was calculated.We also collected concomitant medications known to reduce folic acid (i.e., trimethoprim/sulfamethoxazole, methotrexate, carbamazepine, oxcarbazepine, primidone, valproate, gabapentin, phenytoin, and sulphasalazine) (9)(10)(11) and vitamin B12 levels (i.e., proton pump inhibitor, histamine 2 receptor antagonist, metformin, phenobarbital, pregabalin, primidone, and topiramate) (11)(12) to clarify the association between these medications and macrocytic anemia, which is the main type of olaparib-related anemia (13).Further, we collected data regarding the concomitant use of medications, including strong cytochrome P450 3A4 (CYP3A4) inhibitors (i.e., clarithromycin, erythromycin, itraconazole, ketoconazole, and voriconazole), moderate CYP3A4 inhibitors (i.e., aprepitant, cimetidine, cyclosporin, fluconazole, fluvoxamine, imatinib, posaconazole, and verapamil), and CYP3A4 inducers (i.e., bosentan, carbamazepine, phenytoin, and rifampicin) (14), since olaparib is primarily metabolized by CYP3A4/5 enzymes (84% of total clearance) (15).Although the elevated olaparib plasma exposure was observed in patients with hepatic/renal impairment ( 16), FAERS and JADER did not include clinical laboratory data such as serum creatinine level and alanine aminotransferase.Therefore, we defined the concomitant use of phosphate binders as chronic kidney disease (CKD), since an increase in hyperphosphatemia (serum phosphate level ≥4.5 mg/dl) is notable at an eGFR of about 60 ml/min/1.73m 2 (17,18).Further, we defined the reports of concomitant use of rifaximin and branched chain amino acids as severe liver disease history (19).Since olaparib is one of the therapeutic options available for maintenance therapy of ovarian and pancreatic cancer after platinum-based chemotherapy (20,21), a previous platinum treatment history was also collected.
Adverse events detection.Adverse events were coded according to preferred terms (anemia; 10002034) derived from Medical Dictionary for Regulatory Activities (MedDRA) terminology.Event reports were identified using the standardized MedDRA query (version 25.0).Anemia occurrence during olaparib administration was defined as 'anemia' and other adverse events were defined as 'no anemia'.
Statistical analyses.Statistical analyses were carried out using R software version 4.1.3(R Core Team, 2022) (22).Statistical significance was defined as a two-tailed p <0.05.Categorical and continuous variables were summarized as median [interquartile range (IQR)] and frequency (in percentage), respectively.The FAERS database contains information on both the event-reporting and event-occurring countries, however, the nationalities of the patients are not provided.Therefore, we defined FAERS-Japan as the event-reported country with reference to previous report (7) and conducted sub-group analysis.The association between demographic/clinical factors and olaparib-related anemia was evaluated using reporting odds ratio (ROR) with 95% confidence interval (CI) using univariate logistic regression analysis, similar to previously reported results (23).Adjusted ROR (aROR) for anemia was calculated by adjusting sex, since female is more likely to develop anemia in adulthood (24)(25)(26).In the present study, age, body weight, and DPBW were categorized as binary variables based on the findings of previous reports and median value: age (≥60 years or <60 years, and ≥80 years or <80 years), body weight (≥50 kg or <50 kg), and DPBW (≥12 mg/kg or <12 mg/kg) (27).A statistically significant ROR was formally defined as a lower limit of the 95% CI exceeding 1.0.In FAERS cases, patient characteristics were compared between cases from Japan and other countries using Fisher's exact test.

FAERS.
A flow chart of report selection by FAERS is depicted in Figure 1.A total of 9,685 reports receiving olaparib were extracted, of which 2,981 duplicated reports were excluded from the dataset after data cleaning.The remaining 6,704 reports were subjected to the following analysis.The reports attributed to the occurrence of olaparibrelated anemia were 332 (5%) in FAERS.Among 6,704 total reports, 1,064 (15.9%) were from Japan.
JADER.As shown in Figure 2, there were 1,699 reports in JADER.We conducted data cleaning and excluded 410 duplicated data from the original dataset; the remaining 1,289 reports were subject to the following analysis.The ANTICANCER RESEARCH number of reports of anemia occurrence during the administration of olaparib was 297 (23%).Patient characteristics in JADER are also listed in Table I.There was a significant difference in DPBW between patients with and without anemia [10.9 (10.9-13.3)mg/kg vs. 10.9 (9.2-13.3)mg/kg, p=0.041].

Discussion
Anemia is one of the dose-limiting toxicities of olaparib and has been shown to occur at a higher rate among Japanese ovarian cancer patients than in patients from other countries.Few reports have analyzed differences in the frequency and factors contributing to anemia caused by olaparib by nationality.Therefore, in the present study, we analyzed patient background factors associated with olaparib-related anemia developed in clinical practice using adverse event report databases, FEARS and JADER, at post-marketing stages.Our findings suggested that DPBW has a significant positive association with olaparibrelated anemia, which was demonstrated in both the FAERS and JADER real-world databases.In addition to DPBW, reports from Japan in the FAERS database, concomitant use of drugs causing vitamin B12 deficiency, and previous platinum treatment history were significantly associated with olaparibrelated anemia, A significant difference between reports from Japan and other countries was observed for female sex, body weight, DPBW, and concomitant use of drugs causing vitamin B12 deficiency.Additionally, FAERS-Japan showed that previous platinum treatment history was significantly associated with the anemia occurrence.
Previous physiologically-based pharmacokinetic modeling analysis has revealed that a 50% increase in exposure to olaparib would result in a ~17.5% decrease in hemoglobin level, indicating anemia, in BRCA-mutated ovarian cancer patients (15).Moreover, a recent study revealed an association between olaparib exposure and increased risk of adverse events (28).Since the dose of olaparib is associated with the development of adverse effects, it must be taken under consideration that Japanese people are characterized by a slightly lower body weight as compared to other races (29)(30).In addition, it has been previously demonstrated that underweight Japanese patients had high rates of olaparibrelated adverse events such as severe nausea, lymphopenia, and neutropenia (31).Olaparib is administered at a fixed dose regardless of body weight; therefore, DPBW is relatively high in the population with small body weight.Thus, it is conceivable that Japanese patients are likely to develop anemia due to high DPBW of olaparib.In the same context, it has been reported that underweight individuals have a higher risk of anemia and iron deficiency than overweight female individuals (32)(33).Moreover, it has been revealed that iron deficiency anemia in Japan is higher than that in other countries (24).Nutritional factors may be associated with the anemia occurrence, which differs between nationalities.In the present study, female sex was found to be a significant factor for anemia associated with olaparib in the FAERS database.It is known that females tend to experience anemia owing to menstrual blood loss (34); however, menstruation did not appear to be associated with anemia because the dataset included numerous study subjects at menopause.
Another factor that has been associated with macrocytic anemia is vitamin B12 deficiency, which is common in Western countries (35).Additionally, vitamin B12 deficiency has been reported to be prevalent among underweight people (36).Concomitant use of drugs such as proton pump inhibitors and histamine 2 receptor antagonists has also been shown to block the absorption of vitamin B12 through an increase in gastric pH, causing its deficiency (37).Moreover, a previous study highlighted the relationship between antiepileptic drugs and vitamin B12 deficiency (11).Vitamin B12 is an essential micronutrient in erythropoiesis (38).Long-term administration of antiepileptic drugs causes vitamin B12 deficiency (39), resulting in an increase in MCV and anemia through the alteration of erythropoiesis (38).Olaparib and other drugs causing vitamin B12 deficiency may complementally deteriorate hemoglobin homeostasis, resulting in macrocytic anemia.In our study, olaparib-related anemia was significantly associated with the concomitant use of drugs causing vitamin B12 deficiency only in FAERS, but not in JADER and FAERS-Japan, possibly due to the small number of reports.Generally, it is known that it takes a long-term (e.g., more than 12 months) to develop vitamin B12 deficiency (11,40).We could not investigate the effect of various administration periods of drugs causing vitamin B12 deficiency, which is an issue to be solved in future studies.
Chemotherapy agents can cause DNA damage and olaparib impedes its repair (41,42).Platinum-based agents, such as carboplatin, have also been reported to exert antitumor effects through the formation of DNA adducts (43) and may be associated with an increased risk of anemia (44), which is in agreement with our findings.Therefore, olaparib may contribute to delayed recovery of blood cell counts in patients with history of platinum treatments by inhibiting ANTICANCER RESEARCH 43: 883-891 (2023)  PARP-mediated DNA damage repair.Thus, attention should be paid for anemia occurrence in patients with previous chemotherapy treatments.
There are several limitations of this study.First, spontaneous adverse event reports could not be collected systematically.Although we only define anemia occurrence during olaparib administration as anemia reports, the causality between events and medications is uncertain.Second, this study had incomplete laboratory data.Third, information regarding the concomitant use of drugs that potentially interact with olaparib was lacking from both FAERS and JADER.Fourth, since FAERS and JADER did not include clinical laboratory data such as serum creatinine level and alanine aminotransferase, we could not evaluate renal and liver functions as risk factors of olaparib-related anemia.Finally, the adverse events in FAERS and JADER were reported according to the MedDRA, which did not evaluate the severity of anemia events.

Conclusion
High DPBW was significantly associated with anemia and the concomitant use of drugs causing vitamin B12 deficiency and previous platinum treatment history were shown to contribute to olaparib-related anemia.Understanding the clinical factors in olaparib-related anemia and implementing appropriate follow-up and supportive care for anemia will help in the continuation of olaparib administration.Our findings should be confirmed by registry studies or through further research.

Figure 1 .
Figure 1.Flow chart of the process for data collection of FDA Adverse Event Reporting System (FAERS).

Figure 2 .
Figure 2. Flow chart of the process for data collection of Japanese Adverse Drug Event Report database (JADER).

Table I .
Demographic and clinical characteristics of FAERS and JADER cases.

Table III .
Comparison of demographic and clinical patient characteristics between cases reported from Japan and other countries.

Table IV .
Risk analyses of olaparib-related anemia by FAERS and JADER.

Table V .
Demographic and clinical characteristics of FAERS-Japan cases.

Table VI .
Risk analyses of olaparib-related anemia by FAERS-Japan.Adjusted reporting odds ratio; CI, confidence interval; CKD, chronic kidney disease; DPBW, daily olaparib dose per body weight; FAERS, FDA Adverse Event Reporting System.aROR was adjusted for sex.