The present study reports an incidence of FN at 25% and an incidence of grade 3 or 4 neutropenia at 33% in adults with mCRPC receiving docetaxel. The incidence of FN and neutropenia found in our study was significantly higher than those reported in non-randomized studies with similar study designs.
In a retrospective review by Shigeta et al. (2015), among 95 men with mCRPC receiving docetaxel, FN occurred in 9.5% of patients, with age ≥ 75 years and previous External Beam Radiation Therapy identified as significant predictors for developing FN [12]. However, to meet criteria for inclusion participants had to have adequate bone marrow, liver and kidney function, an absence of other serious medical conditions and could not have received corticosteroid treatment prior to docetaxel. Additionally, inclusion in their analysis was limited to patients who were able to receive docetaxel continuously for more than three cycles regardless if they experienced FN. This is important given in the present study, just over half of the patients in whom FN occurred, developed it after the first cycle and another 20% developed it after either the second or third cycle. Additionally, some of the trial’s patients received treatment every 4 weeks as opposed to the standard every 3 weeks which could have influenced the incidence of FN [12]. Lastly, only nine patients developed FN which limited their ability to detect predictive factors [12].
A study by Kongsted et al. (2016) examined predictors of chemotherapy-induced toxicity among 421 patients receiving docetaxel in combination with low-dose prednisolone for mCRPC. They reported an overall incidence of FN of 11.2% and 12.9% in those aged ≥ 75 years with metastatic epidural spinal cord compression identified as a significant predictor of FN. Similarly, to the present study, this retrospective review did not have strict inclusion criteria and their population had similar baseline characteristics to the present study. However, this study did not report the incidence of neutropenia and did not assess comorbidities as predictive factors for FN risk [11].To our knowledge, the present study is the first to examine the incidence of FN in patients with mCRPC receiving docetaxel in a real-world setting since the implementation of using newer hormonal treatments (i.e. enzalutamide and abiraterone) prior to docetaxel treatment. Both of the previously mentioned retrospective studies took place at a time when it was standard practice to use docetaxel as first-line therapy for metastatic prostate cancer followed by hormone therapies once docetaxel was completed [11, 12]. This does not reflect the change in practice observed over the past 5–10 years which has seen hormone therapies being used first-line and reserving docetaxel for those patients with progression despite hormonal treatment. Therefore, the patients included in the present study are at further lines of therapy and subsequently are further in their disease progression compared to the patients included in previous studies.
According to the ASCO guidelines on the recommendation for use of WBC growth factors, those persons receiving chemotherapy regimens with a > 20% risk of FN should empirically receive G-CSF growth factors as primary prophylaxis [8]. The findings from the present study would suggest all patients with mCRPC receiving docetaxel should receive primary prophylaxis for FN – which would represent a significant change to current practice.
In the present study, belonging to a higher age category and having an increased number of comorbidities were predictors for FN. Advanced age and the presence of multiple comorbidities have been previously reported to be risk factors for FN [8, 10]. Advanced disease has been identified as an independent risk factor for FN [8].
Although there are no existing studies on the efficacy of WBC growth factors specifically in patients with mCRPC receiving docetaxel, an open-label study examining the impact of pegfilgrastim as primary prophylaxis in patients with mCRPC receiving cabazitaxel treatment found a significant reduction in the frequency of FN compared to the rate previously observed without prophylactic G-CSF treatment in a similar patient population (9.5% vs 43%) [6]. Future studies should compare the efficacy of different strategies for secondary prevention of FN, including dose reductions, dose delays and prophylaxis with G-CSF.
This study has some important limitations. The present study design is retrospective however a standardized case report form was used, and data collection was complete with < 1% of data missing. Although a small sample size may have limited the ability to identify other risk factors, with broad inclusion criteria we were able to include almost the entire population of patients with mCRPC receiving docetaxel. Additionally, our 95% CI was narrow enough to indicate that we had enough power to report our primary endpoint with acceptable precision. The current study was limited to a single institution, but practice at The Ottawa Hospital is mostly complicit with published guidelines and we provided enough demographic and clinical data for other centres to compare their practice. Nevertheless, larger multi-centered studies are needed to evaluate the external validity of our study.