Materials and methods
This is a retrospective, single-center, cohort study. The Institutional Review Board of Policlinico Universitario Agostino Gemelli, IRCCS approved the study (ID: 6356). Inclusion criteria were histologically confirmed primary high-grade serous ovarian cancer or fallopian tube carcinoma; FIGO stage III-IV, patients aged ≥ 50 years treated with niraparib between 2019 and 2023 were included. Exclusion criteria were recurrent tubo-ovarian cancer, use of PARP inhibitors as part of an experimental protocol, patients who started niraparib less than 6 months at the time of data analysis and who continued therapy at another center; age < 50 years. To gain a more comprehensive understanding of outcomes in elderly population (≥ 75 years), we stratified them into three groups: group A (50–64 years), group B (65–74 years), and group C (≥ 75 years). The starting dose of niraparib was an individualized starting dose based on weight < or ≥ 77 kg, platelet count < or ≥ 150,000/µL, according to Berek et al. [
10]
. In patients aged 75 years or older, additional considerations for initial dosage were guided by the ECOG (Eastern Cooperative Oncology Group) status, comorbidities and residual toxicity from previous platinum-based chemotherapy. Maintenance treatments in patients aged ≥ 75 years are summarized in Supplementary Fig. 1.
All patients underwent regular monitoring with visits scheduled every 28 days to document toxicities according to the National Cancer Institute Common Terminology Criteria for Adverse Events v5.0 [
11]. Hematological toxicity was closely monitored during the first eight weeks, with weekly blood counts performed. Subsequently, biweekly blood tests were performed, except in cases requiring closer monitoring due to hematologic effects. CT or PET-CT scans was required to evaluate disease progression. The primary outcome was progression-free survival, defined as the time elapsed between the start of treatment with PARP inhibitors and documentation of progressive disease or the date last seen. Secondary outcomes included rates of toxicity and dose reduction.
The whole data are summarized by descriptive statistics measures. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Continuous variables were analyzed using the Kruskall-Wallis test or Mann–Whitney U test. Progression-free survival was estimated using Kaplan–Meier method and compared using the Cox proportional hazards model. The log-rank test was applied to assess statistical significance between survival curves. Logistic regression was used to assess predictors of dose reduction, with results expressed as odds ratios (OR) and 95% confidence intervals (CI). All statistical tests were performed using the Statistical product and Service Solutions software (SPSS, version 29.0; BM Corp., Armonk, NY, United States). Statistical tests were two-sided, and differences were considered significant at the level of p-value < 0.05.
Propensity score matching was performed through R Studio software version 2024.12.1 + 563 (Posit Software, PBC). Patients aged < 75 years and those aged ≥ 75 years were matched according to the following variables: BRCA mutational status, type of surgery (primary debulking surgery versus interval debulking surgery), FIGO disease stage, and residual tumor after cytoreductive surgery.
Discussion
This retrospective real-world study analyzed the safety and efficacy of niraparib in a population of patients aged ≥ 75 years with advanced primary tubo-ovarian cancer. Our results showed that, in older population, niraparib was well tolerated, with a toxicity profile and oncological outcome comparable to younger cohorts. Despite the different baseline characteristics in the ≥ 75-year-old group as higher ECOG value, more frequent use of interval debulking surgery, and a proportion who did not undergo cytoreductive surgery, progression-free survival did not significantly differ from those observed in younger patients (p = 0.78). Moreover, dose reductions occurred less frequently in the elderly compared to younger patients, although this difference was not statistically significant (p = 0.08). Elderly patients exhibited a slightly higher incidence of grade ≥ 3 thrombocytopenia and no significant increase in non-hematologic or other severe adverse events was observed.
In the context of PARP inhibitors maintenance therapy, clinical trials have enrolled only a limited percentage of patients aged 65 and older: SOLO1 (13.8%), PAOLA-1 (36.2%), VELIA (39.7%), and PRIMA (39.4%) [
2‐
4,
12]. This limited representation restricts the applicability of clinical trial data to real-world elderly populations. Recently, a meta-analysis by Maiorana et al. examined the efficacy of PARP inhibitors in elderly patients and concluded that age does not significantly affect the clinical benefits of PARP inhibitors [
13]. This reinforces the rationale for the effective and routine use of PARP inhibitors in older populations.
Moreover, residual tumor following debulking surgery has been reported to be higher in elderly patients, representing the main prognostic factor for survival outcomes [
14‐
18]. When providing suboptimal treatment to the elderly because of their age, numerous studies focused on primary treatment and demonstrated shorter survival outcome [
19]. Warren et al., reported that only 37.6% of patients aged over 75 years underwent appropriate surgery, while Cloven et al. showed that optimal cytoreduction was achieved in only 29.5% of patients aged 60–79 years, decreasing further to 21.7% among those older than 80 years [
17,
18]. Conversely, Joueidi et al
. reported residual disease following cytoreduction surgery in 30% of patients aged ≥ 75 years, compared to 20% in those under 65 years (
p = 0.04), with 70% of elderly patients achieving complete cytoreduction [
16]. Similarly, in our cohort, a residual tumor > 1 cm or the inability to achieve optimal cytoreduction despite neoadjuvant chemotherapy was higher in group C (30.8%) compared to groups A and B, in which all patients achieved either no residual disease or a residual tumor ≤ 1 cm. Despite these challenges, the efficacy of niraparib was not compromised in our elderly population. This could suggest that advanced age should not preclude patients from receiving niraparib, as its therapeutic benefits remain consistent across age groups.
We noticed that severe thrombocytopenia was more frequent in group C (33.3%) compared to groups A and B (24.2% and 15.4%, respectively). This finding may be attributed to the more frequent use of medications such as anticoagulants, antiplatelet agents or to nutritional deficiencies (e.g., vitamin B12, folate) in the elderly population [
20‐
23]. This is consistent with the ≥ 75 years subgroup of the PRIMA trial, which reported a severe thrombocytopenia rate of 35.3% in patients treated with the individualized starting dose, while the rate remained significantly higher (62.2%) in those who received the fixed standard dose [
2].
Conversely, we found that the older population had a lower risk of grade 3–4 anemia (12.8%) compared to group A (28.3%). Our finding is also supported by the meta-analysis by Maiorana et al., which reported that older patients seem to have a lower risk of developing severe anemia [
13].
Nonetheless, it should be underlined that 23.1% of patients in our elderly cohort started with an initial dose of 100 mg/day compared to only 1.1% of patients under 75 years.
A higher rate of dose reduction was observed in groups A and B (77.4% and 69.2%, respectively) compared to group C (56.4%); however, this difference was not statistically significant (
p = 0.08). This slight difference may be attributable to the more frequent use of individualized starting doses among patients aged ≥ 75 years (100 mg/day in 23.1% of the older group vs 1.1% in the younger groups), as well as to the consequently different starting doses of niraparib across these groups (higher in groups A and B) (Table
2). When the analysis was restricted through propensity score matching to 39 patients per age group (< 75 and ≥ 75 years), the difference in dose reduction between the matched groups became statistically significant (
p < 0.001), supporting the influence of the starting dose on subsequent dose reductions.
However, it should also be emphasized that, even when beginning with lower doses, older patients remain more susceptible to thrombocytopenia than to anemia. This suggests that the drug's toxicity is only partly dependent on dosage; it is also related to age and the characteristics of individual patients.
Interestingly, the dose reduction rate observed in elderly patients in our series was lower than the 61.5% reported in the PRIMA trial [
2]. This difference is probably due to the use of the individualized starting dose and a more tailored dosing strategy in the elderly group, which considered general clinical conditions, comorbidities, residual post-chemotherapy toxicity, and age.
No substantial differences were observed in our cohort between the elderly and younger populations regarding non-severe hematologic adverse events and the incidence of severe neutropenia (Fig.
3). The overlapping trends observed in the toxicity curves reinforce the hypothesis that age alone does not substantially impact the overall safety profile of niraparib. These finding are consistent with data reported in previous clinical trials [
2,
8].
No statistically significant differences were observed between groups in terms of dose discontinuation (
p = 0.12). Moreover, the percentage of patients aged ≥ 75 years who discontinued niraparib due to grade ≥ 3 adverse events (17.9%) was comparable to the rate reported in the same population in the PRIMA trial (18.5%) [
14]. This suggests that, with appropriate clinical management, the majority of older patients could continue therapy without significant interruptions. However, the occurrence of rare but severe events such as myelodysplastic syndrome and acute myeloid leukemia emphasizes the importance of continuous clinical monitoring, particularly in older patients who may already have compromised hematologic reserves.
Our data also suggest that older patients have comparable short-term survival outcomes to younger patients (HR 0.86, 95%CI: 0.45–2.41;
p = 0.37), indicating that chronological age alone is not an independent predictive factor in tube-ovarian cancer. This is further supported by findings from the sub-analysis of IMPRESSS study, which enrolled 721 patients with epithelial ovarian cancer, including 226 (31%) patients aged ≥ 75 years. In this cohort, older patients who received treatment similar to that of younger women demonstrated similar survival rates. Notably, there was no association between age at diagnosis and progression free survival (
p = 0.40), instead outcomes appeared to be primarily influenced by the type of treatment received [
24].
We recognize that the retrospective design of the study and the lack of available quality of life data represent limitations in the interpretation of these findings. Additionally, the relatively small sample size may limit statistical power, and differences in baseline characteristics across the groups may have introduced potential confounding. Our findings should be interpreted considering the significant baseline differences observed between the ≥ 75-year group and the younger cohorts, particularly in ECOG performance status, surgical approach, and residual disease status. These imbalances may have introduced residual confounding in the progression-free survival analysis, independent of age or treatment-related variables. While propensity score-based methods are recommended to mitigate such biases, the relatively small sample size in our elderly subgroup would have resulted in substantial loss of statistical power and reduced interpretability of matched analyses. Consequently, our results should be interpreted as hypothesis-generating and confirmatory analyses using larger multi-center datasets with propensity-based adjustments are warranted.
Nonetheless, this is the largest real-world study investigating the management of very elderly patients treated with PARP inhibitors. This underscores the importance of tailoring treatment strategies in this population, including the individualized starting doses of niraparib.
As life expectancy continues to rise, the incidence of tubo-ovarian cancer in the elderly is expected to increase substantially. Consequently, oncologic management in this population is emerging as a critical issue for clinicians. Real-world data often provide a more accurate representation of older patients with various comorbidities and highlight the relevance of dose modifications and proactive management of adverse events to optimize treatment outcomes. These patients, particularly those aged ≥ 75 years, are often underrepresented in clinical trials, although the median age at tubo-ovarian cancer diagnosis is 63 years [
1]. Our results reinforce the concept that age alone should not be considered a limiting factor when prescribing niraparib. With individualized starting doses and proactive monitoring strategies, older patients can benefit from maintenance therapy with manageable toxicity. These findings suggest that clinicians should consider incorporating elderly patients into maintenance regimens more systematically, avoiding age-based exclusions. Furthermore, future clinical trials should strive to include a representative proportion of older patients to enhance both external validity and equity in cancer care. Additional prospective studies, ideally including geriatric assessment tools, are warranted to refine treatment selection and enhance outcomes in this vulnerable population.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.