Welcome, this website is intended for international healthcare professionals with an interest in the treatment of Advanced Prostate Cancer. By clicking the link below you are declaring and confirming that you are a healthcare professional

You are here

Severe neutropenia during cabazitaxel treatment is associated with survival benefit in men with metastatic castration-resistant prostate cancer (mCRPC): A post-hoc analysis of the TROPIC phase III trial

European Journal of Cancer, March 2016, Pages 93 - 100

Abstract

Background

Cabazitaxel significantly improves overall survival (OS) in men with metastatic castration-resistant prostate cancer (mCRPC) progressing during or after docetaxel, but is associated with a higher rate of grade ≥3 neutropenia compared with docetaxel. We thus examined the relationship between cabazitaxel-induced grade ≥3 neutropenia, baseline neutrophil-lymphocyte ratio (NLR) and treatment outcomes.

Methods

Data from the experimental arm of the TROPIC phase 3 trial which randomly assigned men with mCRPC to cabazitaxel or mitoxantrone every 3 weeks, both combined with daily prednisone, were analysed. The influence on OS (primary end-point) and progression-free survival (PFS) of at least one episode of grade ≥3 neutropenia during cabazitaxel therapy was investigated using Cox regression models, adjusted for pain at baseline. The relationships with prostate-specific antigen (PSA) responses during cabazitaxel therapy and baseline NLR were also analysed.

Findings

The occurrence of grade ≥3 neutropenia during cabazitaxel therapy was associated with a prolonged OS (median 16.3 versus 14.0 months, hazard ratio (HR) [95% confidence interval] = 0.65 [0.43–0.97], p = 0.035), a twice longer PFS (median 5.3 versus 2.6 months, HR = 0.56 [0.40–0.79], p = 0.001) and a higher confirmed PSA response ≥50% (49.8% versus 24.4%, p = 0.005), as compared with patients who did not develop grade ≥3 neutropenia. Grade ≥3 neutropenia was more common in case of NLR <3 as compared with NLR ≥3 at baseline (88.8% versus 75.3%, p = 0.002). Combining low NLR at baseline and grade ≥3 neutropenia during therapy was associated with the longest OS (median 19.2 months) while high NLR at baseline and no grade ≥3 neutropenia was associated with a poor OS (median 12.9 months, HR 0.46 [0.28–0.76], p = 0.002). In the subgroup of neutropenic patients the median OS was 19.7 months in those treated with granulocyte colony-stimulating factor (G-CSF) and 16 months on those without G-CSF support.

Interpretation

This post-hoc analysis of TROPIC suggests that the occurrence of grade ≥3 neutropenia with cabazitaxel is associated with improved OS and PFS. Patients with a low NLR at baseline were more likely to develop grade ≥3 neutropenia during cabazitaxel therapy and showed the longest OS. High NLR at baseline and no grade ≥3 neutropenia during therapy was associated with poor outcomes which may suggest insufficient drug exposure or a limited impact on the tumour-associated immune response. Primary or secondary prophylactic use of G-CSF had no adverse impact for outcome. If prospectively confirmed, these results would justify maintaining the intended cabazitaxel dose of 25 mg/m2 whenever possible.

Highlights

  • Cabazitaxel is an option in metastasized castration resistant prostate cancer (mCRPC).
  • The occurrence of cabazitaxel induced neutropenia (CIN) is a positive predictive factor.
  • Severe CIN is associated with better progression-free and overall survival.
  • Prophylactic use of filgrastim (G-CSF) after first CIN shows a survival advantage.
  • A model of neutrophil-to-lymphocyte ratio and CIN defines three risk categories in mCRPC.

Keywords: TROPIC, Post hoc analysis, Cabazitaxel chemotherapy, Metastasized castration resistent prostate cancer, Neutropenia predictive, Granulocyte-colony stimulating factor (GCSF), mCRPC.

1. Introduction

Cabazitaxel was the first agent demonstrating a survival benefit in men with metastatic castration-resistant prostate cancer (mCRPC) progressing during or after docetaxel [1]. Although cabazitaxel is associated with less peripheral neuropathy, nail disorders, alopecia and dysgeusia than docetaxel [2], it induces a high rate of grade ≥3 neutropenia which needs to be proactively managed to avoid potential neutropenic complications [3]. This contributes to limit its use compared with new androgen receptor (AR)-targeted agents such as abiraterone acetate [4] or enzalutamide [5] which are given orally with less adverse events. Nevertheless, the increasing evidence that prostate cancer is a highly heterogeneous disease [6] and [7], that some patients exhibit primary resistance to AR-targeted agents [8] and [9] and that responders to AR-targeted agents will ultimately progress through various mechanisms such as splice variants [10] or aggressive histological subtypes which are not driven by AR [11], accords cabazitaxel an important role in such a setting.

In many solid tumour types and haematological malignancies, it has been reported that chemotherapy-induced neutropenia is associated with a prolonged overall survival (OS), both in adjuvant and metastatic settings [12], [13], [14], [15], [16], [17], [18], [19], and [20]. In mCRPC, in a retrospective review of 221 patients treated with docetaxel, Pond et al [21] concluded that development of a grade ≥3 neutropenia on day 8 of cycle 1 was associated with a better outcome. Since a particularly high incidence of grade ≥3 neutropenia (82%) was reported with cabazitaxel, administered at a dose of 25 mg/m2 every 3 weeks, in the TROPIC trial [1], we examined whether this was associated with improved treatment outcome. We also investigated grade ≥3 neutropenia in relation to the neutrophil-lymphocyte ratio (NLR). The NLR is the quotient of baseline absolute peripheral neutrophils (cells/mm3) and baseline absolute lymphocytes (cells/mm3) and has been shown to have prognostic value in mCRPC [22], [23], and [24] as in many other tumour types [25]. Lorente et al. have demonstrated an association of a high NLR with decreased survival, PSA and Response Evaluation Criteria in Solid Tumours responses in patients treated with cabazitaxel [22]. A decrease of the NLR from high (≥3) to low (<3) resulted in improved survival (hazard ratio [HR] 0.66; 95% confidence interval [CI] 0.51–0.85; p = 0.001) and better PSA response rates (66.4% versus 33.6%; p = 0.000) [22]. Finally, we investigated the relationship between grade ≥3 neutropenia, granulocyte colony-stimulating factor (G-CSF) usage and OS.

2. Methods

2.1. Patients

We conducted a post-hoc analysis of the randomised phase III trial TROPIC which compared the efficacy of cabazitaxel (25 mg/m2 every 3 weeks) versus mitoxantrone (12 mg/m2 every 3 weeks), both in combination with prednisone 10 mg daily, in 755 patients with mCRPC progressing during or after docetaxel. Full details of patient eligibility and trial results have been published [1]. The study was approved by the local ethic committee (EKNZ UBE 15/105).

Patient characteristics collected at enrolment, included number and duration of prior lines of treatment, sites of metastases, age, Eastern Cooperative Group performance status, pain score, testosterone and PSA concentrations, steroid use, full blood count (including absolute neutrophil and lymphocyte counts) and biochemistry. Full blood counts were carried out on a weekly basis (day 1, 8 and 15 of each 21-day cycle). Imaging studies (computed tomography and bone scintigraphy) were carried out every 12 weeks. Treatment was planned for a maximum of 10 cycles. Prophylactic G-CSF was not allowed during the first cycle, but was allowed (at physicians' discretion) after first occurrence of either a grade ≥3 neutropenia lasting 7 d or more or a grade ≥3 neutropenia complicated by fever or infection.

2.2. Statistical analyses

All analyses reported below were performed in those 371 patients exposed to cabazitaxel in the TROPIC trial.

The primary objective of this study was to explore the prognostic role of grade ≥3 neutropenia during cabazitaxel therapy on OS, calculated from date of randomisation to death. Secondary objectives were to evaluate the effect of grade ≥3 neutropenia on progression-free survival (PFS), defined as the time between randomisation and the first date of progression (either PSA progression or tumour progression or pain progression or death) and PSA response, defined by a confirmed PSA decrease ≥50% from baseline.

OS and PFS were estimated by Kaplan-Meier method and comparisons were made by the log rank test. We investigated using Cox proportional-hazards survival models whether (A) the occurrence of at least one grade ≥3 neutropenia at any time during treatment, (B) the frequency of grade ≥3 neutropenia during therapy (i.e. number of cycles with grade ≥3 neutropenia divided by the total number of cycles received), (C) the relative change in absolute neutrophil count (ANC) during treatment cycle 1 compared to baseline and (D) the time to first grade ≥3 neutropenia (as continuous variable) had an influence on OS and PFS with cabazitaxel. The variable ‘measurable pain at baseline’ (1 = yes, 0 = no) was included in the model. For analyses related to variables (A) and (B), only patients who completed at least three treatment cycles without missing covariates were included (n = 259). The Wilcoxon signed rank sum test was employed to test whether the occurrence of grade ≥3 neutropenia (A) and the relative change in ANC during treatment cycle 1 (C) were associated with a confirmed PSA decrease ≥50%. Waterfall plots of maximum PSA changes from baseline during therapy in patients experiencing or not at least one grade ≥3 neutropenia were also provided.

The NLR was calculated by dividing the ANC at baseline by the absolute lymphocytes count (cells/mm3). This continuous variable was further stratified into high NLR (≥3.0) and low NLR (<3.0) according to the suggestion of Lorente et al [22]. Based on this, we defined four groups of patients: (1) grade ≥3 neutropenia and high NLR, (2) grade ≥3 neutropenia and low NLR, (3) no grade ≥3 neutropenia and high NLR, and (4) no grade ≥3 neutropenia and low NLR. We tested the two groups ‘high NLR’ and ‘low NLR’ for a difference in proportions of patients with and without grade ≥3 neutropenia, using a two-sample test for equality of proportions with continuity correction. OS in each of these subgroups was analysed using a Cox proportional-hazards survival model, adjusted for measurable pain at baseline. We also tested the association between NLR (as a continuous variable) and OS.

Primary prophylaxis with G-CSF was not permitted at cycle 1; however, as per protocol, G-CSF use was permitted for primary prophylaxis in case of first occurrence of either a grade ≥3 neutropenia lasting 7 d or more or a grade ≥3 neutropenia complicated by fever or infection. We further grouped patients with grade ≥3 neutropenia according to G-CSF use (none, after first grade ≥3 neutropenia, before first grade ≥3 neutropenia), and compared their OS to those patients without grade ≥3 neutropenia, using Cox proportional-hazards survival model, adjusted for measurable pain at baseline.

3. Results

3.1. Primary end-point: OS

Patients developing grade ≥3 neutropenia during cabazitaxel therapy had a significantly prolonged OS compared to those without grade ≥3 neutropenia (16.3 versus 14.0 months; HR = 0.65 [95% CI, 0.43–0.97], p = 0.035) (Fig. 1A).

gr1

Fig. 1 A: Overall survival (OS) according to the occurrence of at least one grade ≥3 neutropenia during cabazitaxel therapy. Men with at least one grade ≥3 neutropenia during therapy (n = 254) had a median OS of 16.3 months compared to 14.0 months in those without grade ≥3 neutropenia (n = 46; HR = 0.65 [0.42–0.97], p = 0.035). Vertical lines indicate median OS. Only patients exposed to a minimum of three cycles of cabazitaxel were included in this analysis. B: Progression-free survival (PFS) according to the occurrence of grade ≥3 neutropenia. Patients with at least one episode of grade ≥3 neutropenia (n = 256) had a median PFS of 4.6 months compared to 2.1 months in those without grade ≥3 neutropenia (n = 67, HR = 0.56 [0.40–0.79], p = 0.001). Vertical lines indicate median PFS. Only patients exposed to a minimum of three cycles of cabazitaxel were included in this analysis.

There was also a trend for improved OS in patients experiencing a high frequency of grade ≥3 neutropenia during therapy (HR = 0.61 [0.36–1.04], p = 0.067).

OS was not affected by the relative change in ANC from baseline during the first cycle (HR = 0.97 [0.57–1.66], p = 0.908) and by the time to first grade ≥3 neutropenia (HR = 0.99 [0.97–1.01], p = 0.404).

Of the total of 303 patients that experienced at least grade 3 neutropenia during treatment 242 (79.9%) had neutropenia in their first cycle. A total of 274 patients (90.5%) had neutropenia within the first two cycles and 288 (95.1%) patients had neutropenia within their first three cycles of cabazitaxel.

3.2. Secondary end-point: PFS

Patients with grade ≥3 neutropenia during cabazitaxel therapy had a longer PFS compared to those with a lower grade neutropenia (5.3 versus 2.6 months; HR = 0.56 [0.40–0.79], p = 0.001) (Fig. 1B).

PFS also significantly increased with the frequency of grade ≥3 neutropenia during therapy (HR = 0.63 [0.42–0.95], p = 0.029). A beneficial effect on PFS was already apparent after one single episode of grade ≥3 neutropenia.

Lastly, patients with small relative changes from baseline in ANC during cycle 1 tended to have a reduced PFS (HR = 1.49 [0.98–2.25], p = 0.061).

3.3. Secondary end-point: PSA response

Patients with grade ≥3 neutropenia during cabazitaxel therapy had a greater fall in PSA from baseline compared with cases of milder neutropenia (median –48.4 versus –4.6, p < 0.001). Only 10/41 patients (24.4%) who did not experience grade ≥3 neutropenia had a confirmed PSA decline ≥50% compared to 104/209 patients (49.8%) with grade ≥3 neutropenia (p = 0.005) (Fig. 2 and Supplementary Fig. 1).

3.4. Analysis of the NLR in combination with grade ≥3 neutropenia

Of the 371 patients exposed to cabazitaxel, 198 had a NLR ≥3 (high NLR), 169 had an NLR <3 (low NLR), and four patients had missing NLR values. The occurrence of grade ≥3 neutropenia during cabazitaxel therapy was significantly lower in patients with a high NLR (149/198, 75.3%) compared with patients with a low NLR (150/169, 88.8%, p = 0.002).

It was previously reported that patients with high NLR had a worse OS compared to those with low NLR (12.6 versus 15.9 months, p < 0.001) [22]. Combined evaluation of the occurrence of grade ≥3 neutropenia and NLR, showed that patients with a low baseline NLR who developed grade ≥3 neutropenia during therapy had the longest OS (19.2 months), while those with a high NLR and no grade ≥3 neutropenia during therapy had the worst OS (12.9 months, HR 0.46 [95% CI 0.28–0.76], p = 0.002)

3.5. A risk model based on NLR and occurrence of grade ≥3 neutropenia

Since high NLR at baseline and absence of grade ≥3 neutropenia during therapy were associated with a poor outcome, we developed a risk model integrating these two factors. Patients were classified into three groups: a low-risk group (low NLR at baseline, grade ≥3 neutropenia during cabazitaxel therapy), a high-risk group (high NLR at baseline, no grade ≥3 neutropenia during therapy) and an intermediate-risk group (only one risk factor, either high NLR or no grade ≥3 neutropenia). The median OS in the low-risk group (n = 130) was 19.2 months and significantly longer when compared with 12.9 months in the high-risk (n = 35), (p = 0.0024) and 14.2 months in the intermediate-risk group (n = 132), (p = 0.0024) (Fig. 3).

3.6. G-CSF use

Of the 371 patients analysed, 303 (82%) experienced at least one episode of grade ≥3 neutropenia during cabazitaxel therapy and 77/303 (21%) received G-CSF prophylaxis, either after a prior grade ≥3 neutropenia (n = 59) or not (n = 18). Of those 77 patients treated with G-CSF, 76 (99%) subsequently developed at least one episode of grade ≥3 neutropenia versus 266/294 (77%) who did not receive G-CSF. Median OS was 19.7 months in patients treated with prophylactic G-CSF after a first episode of neutropenia grade ≥3 (n = 59), 17.3 months in those treated with G-CSF without prior grade ≥3 neutropenia and 16.0 months in those who never received G-CSF (Fig. 4 and Supplementary Table 2).

gr2

Fig. 2 PSA response according to the occurrence of grade ≥3 neutropenia with cabazitaxel. Waterfall plot showing maximum PSA change from baseline according to whether patients developed at least one episode of grade ≥3 neutropenia during cabazitaxel therapy (blue bars, n = 259) or not (red bars, n = 57). PSA response was defined as a confirmed ≥50% decrease from baseline (dotted line). The y-axis is capped at +100% for better representation (maximum PSA rise was 377%). Patients who experienced at least one grade ≥3 neutropenia had greater PSA declines than those without grade ≥3 neutropenia.

gr3

Fig. 3 Overall survival (OS) according to risk groups (based on neutrophil-to-lymphocyte ratio [NLR] and occurrence of grade ≥3 neutropenia). The curves show the probability of survival (in months) by number of risk groups: 0 (at least one grade ≥3 neutropenia during cabazitaxel therapy and low NLR), 1 (high NLR or no grade ≥3 neutropenia during cabazitaxel therapy) 2 (high NLR and no grade ≥3 neutropenia during cabazitaxel therapy). Vertical lines indicate median OS. Only patients who completed at least three treatment cycles were included. Comparison of outcome between the risk groups is depicted in Table 1.

gr4

Fig. 4 Outcome (OS) of patients with and without secondary prophylactic G-CSF use. Kaplan-Meier estimates of the probability of overall survival stratified for neutropenia grade ≥3 and G-CSF use after a first episode of neutropenia grade ≥3. Vertical lines indicate median overall survival in months. Patients without neutropenia grade ≥3 were not allowed G-CSF according to the protocol. One patient who received G-CSF without preceding neutropenia grade ≥3 was excluded from the analysis. All patients who completed at least three treatment cycles and who did not receive G-CSF before first episode of neutropenia grade ≥3 were included (n = 204). Eighteen patients who received G-CSF prophylaxis before their first neutropenic episode had a median OS of 17.3 months and are not displayed in this figure. G-CSF, granulocyte colony-stimulating factor.

Table 1 OS stratified according to risk groups.

HR 95% CI z p (>|z|)
Risk group 1 versus 0 1.72 [1.21; 2.44] –3.03 0.00241
Risk group 2 versus 0 2.18 [1.32; 3.61] –3.03 0.00241
Pain at baseline 2.40 [1.71; 3.36] 5.09 <0.001

OS, overall survival; HR, hazard ratio; CI, confidence interval.

4. Discussion

This post-hoc analysis of TROPIC suggests that the occurrence of grade ≥3 neutropenia with cabazitaxel is associated with improved OS but also greater PFS and PSA responses. Patients with a low NLR at baseline were more likely to develop grade ≥3 neutropenia with cabazitaxel therapy and showed the longest OS (median 19.2 months). Conversely, high NLR at baseline and no grade ≥3 neutropenia during therapy was associated with a poor outcome (median OS of 12.9 months). Importantly, all patients had a full blood count on day 8, 15 and 21 enabling the capture of the neutrophil nadir during therapy. The majority (90.5%) of neutropenic events occurred early, within the first two cycles of therapy, making a cumulative effect unlikely.

Chemotherapy-induced neutropenia is associated with an improved OS in many solid tumours such as breast cancer [19], colorectal cancer [23] and [25], non-small-cell lung cancer [17], gastric cancer [16], cervical cancer [15], nasopharyngal cancer [13] and haematological malignancies [12] and [26]. In the context of mCRPC, one retrospective analysis in 221 patients treated with docetaxel (75 mg/m2 every 3 weeks) also suggests that the occurrence of grade ≥3 neutropenia on day 8 of cycle 1 is associated with an improved OS and a greater PSA response compared with those without grade ≥3 neutropenia [21]. Our results further support these findings as cabazitaxel-induced grade ≥3 neutropenia in the TROPIC study is also associated with a better outcome.

The large scale TAX327 study, demonstrated a survival advantage (median 2.4 months) for docetaxel plus prednisone versus mitoxantrone plus prednisone in mCRPC [27]. The incidence of grade ≥3 neutropenia and febrile neutropenia were 32% and 3%, respectively. Several clinical studies were then initiated at the same dose in earlier stages of the disease (i.e. in hormone-naive patients) and showed a higher incidence of grade ≥3 neutropenia (61–88%) [28]. The lower rate of grade ≥3 neutropenia in mCRPC was attributed to a dramatic increase in docetaxel clearance (by approximately 100%) due to increased hepatic uptake compared to hormone-naive men, resulting in a two-fold reduction in the area under the curve [28]. Results of two large phase III trials, CHAARTED (n = 790) and STAMPEDE (n = 1776), have been reported [29] and [30]. These trials compared docetaxel (six cycles) plus androgen deprivation therapy (ADT) versus ADT alone in metastatic hormone-naive patients. A third trial GETUG 15 which included only 385 hormone-naive metastatic patients showed a significant PFS improvement with the combination of ADT and docetaxel but only a trend in OS [31]. Interestingly, a much higher rate of febrile neutropenia was reported in these three studies (ranging from 6% to 12%) compared with TAX327 (3%) reflecting possibly a greater incidence of grade ≥3 neutropenia than effectively reported.

In the TROPIC study, conducted in mCRPC patients progressing during or after docetaxel, the incidence of grade ≥3 neutropenia and febrile neutropenia with cabazitaxel (25 mg/m2 every 3 weeks) was much higher (82% and 8%, respectively) than with docetaxel in the TAX327 study (32% and 3%, respectively), although the OS benefit versus mitoxantrone plus prednisone was similar (median 2.4 months) [1]. A large compassionate use program providing access to cabazitaxel at a dose of 25 mg/m2 to 746 mCRPC patients progressing during or after docetaxel reported 17% of grade ≥3 neutropenia and 5.5% of febrile neutropenia [32]. In comparison with TROPIC the improved tolerability was interpreted to be due to better monitoring of patients and proactive management of adverse events, including use of prophylactic G-CSF in 52% of patients. Indeed, a multivariate analysis concluded that age ≥75 years, treatment cycle 1, and neutrophil count <4000/mm3 before cabazitaxel injection were associated with increased risk of developing grade ≥3 neutropenia and/or neutropenic complications but in the presence of these factors, prophylactic G-CSF at a given cycle significantly reduced this risk by 30% (odds ratio 0.70 [0.49–0.99], p = 0.04) [32]. Two phase III studies (NCT01308580 – PROSELICA in post-docetaxel setting and NCT01308567 – FIRSTANA in chemo-naive patients) are ongoing and should unambiguously determine if a dose of 20 mg/m2 provides a greater benefit/risk ratio than 25 mg/m2. Pharmacokinetic (PK) guided dose individualisation of cabazitaxel is currently tested versus the standard regimen (25 mg/m2 every 3 weeks) in the prospective randomised phase II study CAINTA (EudraCT number: 2013-005504-34).

Thus, if validated prospectively, a practical implication of our results might be to tailor cabazitaxel dosing on the basis of its haematological effects. Indeed, the absence of grade ≥3 neutropenia, especially in patients with a high NLR, could suggest insufficient drug exposure or limited impact on the tumour-associated immune response. Our data indicate that G-CSF use after successful dose calibration could further positively impact patient survival. A risk stratification according to our model could be helpful in selecting patients who would eventually benefit from dose escalation, provided all safety measures are considered.

In conclusion, this post-hoc analysis of the TROPIC trial suggests that the occurrence of grade ≥3 neutropenia with cabazitaxel is associated with a prolonged OS and doubles the PFS. Patients with a low NLR at baseline were more likely to develop grade ≥3 neutropenia during cabazitaxel therapy and showed the longest OS. High NLRs at baseline and no grade ≥3 neutropenia during therapy was associated with a poor outcome which may suggest under dosing. These results, if prospectively confirmed, would justify keeping the cabazitaxel dose at 25 mg/m2 whenever possible.

Conflict of interest statement

AM and FS have served as consultants for Sanofi and have received funds for travelling from Sanofi.

SvF, DV and HL have declared to have no conflict of interests.

RdW, OS and JdB have served as investigator and consultant for Sanofi.

Acknowledgment

We thank Ms. Gillian von Rechenberg for critical reading the manuscript and helpful editorial input.

Appendix A. Supplementary data

The following are the supplementary data related to this article:

figs1

Supplementary Fig. 1 Waterfall plots for PSA changes from baseline according to whether patients experienced at least one episode of grade ≥3 neutropenia during cabazitaxel therapy or not, for patients with high NLR (>3; S1a) and low NLR (<3; S1b). Confirmed PSA decrease ≥50% was less frequent in patients with high NLR as compared to those with low NLR (54/136 (39.7%) versus 59/111 (53.2%), p = 0.048). The y-axis is capped at +100% for better representation. Only patients who completed at least three treatment cycles were included; 22 patients with a non-confirmed PSA response were excluded. NLR, neutrophil-to-lymphocyte ratio.

figs2
figs3

Supplementary Fig. 2 Kaplan-Meier estimates of the probability of overall survival (OS) (in months) according to the occurrence or not of at least one episode of grade ≥3 neutropenia and NLR (<3 versus ≥3). Vertical lines indicate median OS. Red line: patients without grade ≥3 neutropenia and high NLR (n = 35); dark blue line: patients with at least one episode of grade ≥3 neutropenia and high NLR (n = 121); light blue line: patients with at least one episode of grade ≥3 neutropenia and low NLR (n = 130); yellow line: patients without grade ≥3 neutropenia and low NLR (n = 11). Only patients exposed to a minimum of three cycles cabazitaxel were included in this analysis. Statistical details are shown in Supplementary Table 1. NLR, neutrophil-to-lymphocyte ratio.

Download file

Supplementary Table 1 The table shows the respective statistics of Fig. 3. Hazard ratio (HR) estimates from the Cox model for overall survival stratified according to the occurrence of grade ≥3 neutropenia (NP: at least one episode; no-NP: no grade ≥3 neutropenia) and NLR (high, ≥3; low, <3). Contrast options were set to treatment contrast with the highest risk group (no grade ≥3 neutropenia, high NLR) as reference group. The column ‘95% CI’ gives the limits of the 95% confidence intervals for the HR. A significant difference is indicated if the limits do not include 1 (HR different from 1). Only patients who completed at least three treatment cycles without missing covariates were included (n = 256). NLR, neutrophil-to-lymphocyte ratio.

Download file

Supplementary Table 2 Hazard ratio (HR) estimates from the Cox model for overall survival, stratified for neutropenia grade 3/4 and G-CSF use. The table shows the respective statistics of Fig. 4. The column ‘95% CI’ gives the limits of the 95% confidence intervals for the HR. A significant difference is indicated if the limits do not include 1 (HR different from 1). Only patients who completed at least three treatment cycles without missing covariates were included (n = 244). NP, neutropenia; G-CSF, granulocyte colony-stimulating factor; ECOG, Eastern Cooperative Group.

References

  • [1] J.S. de Bono, S. Oudard, M. Ozguroglu, S. Hansen, J.-P. Machiels, I. Kocak, et al. Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after docetaxel treatment: a randomised open-label trial. Lancet. 2010;376:1147-1154 10.1016/S0140-6736(10)61389-X Crossref
  • [2] A. Omlin, O. Sartor, C. Rothermundt, R. Cathomas, J.S. De Bono, L. Shen, et al. Analysis of side effect profile of alopecia, nail changes, peripheral neuropathy, and dysgeusia in prostate cancer patients treated with docetaxel and cabazitaxel. Clin Genitourin Cancer. 2015;13:e205-e208 10.1016/j.clgc.2015.01.010 Crossref
  • [3] A. Heidenreich, H.-J. Scholz, S. Rogenhofer, C. Arsov, M. Retz, S.C. Müller, et al. Cabazitaxel plus prednisone for metastatic castration-resistant prostate cancer progressing after docetaxel : results from the German compassionate-use programme. Eur Urol. 2013;63:977-982 10.1016/j.eururo.2012.08.058 Crossref
  • [4] J.S. De Bono, C.J. Logothetis, A. Molina, K. Fizazi, S. North, L. Chu, et al. Abiraterone and Increased survival in mPrCa. NEJM. 2011;364:1995-2005 Crossref
  • [5] H.I. Scher, K. Fizazi, F. Saad, M.-E. Taplin, C.N. Sternberg, K. Miller, et al. Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med. 2012;367:1187-1197 10.1056/NEJMoa1207506
  • [6] D. Robinson, E.M. Van Allen, Y.-M. Wu, N. Schultz, R.J. Lonigro, J.-M. Mosquera, et al. Integrative clinical genomics of advanced prostate cancer. Cell. 2015;161:1215-1228 10.1016/j.cell.2015.05.001 Crossref
  • [7] G. Gundem, P. Van Loo, B. Kremeyer, L.B. Alexandrov, J.M.C. Tubio, E. Papaemmanuil, et al. The evolutionary history of lethal metastatic prostate cancer. Nature. 2015;520:353-357 10.1038/nature14347 Crossref
  • [8] E. Efstathiou, M. Titus, S. Wen, A. Hoang, M. Karlou, R. Ashe, et al. Molecular characterization of enzalutamide-treated bone metastatic castration-resistant prostate cancer. Eur Urol. 2015;67:53-60 10.1016/j.eururo.2014.05.005 Crossref
  • [9] E. Efstathiou, M. Titus, D. Tsavachidou, V. Tzelepi, S. Wen, A. Hoang, et al. Effects of abiraterone acetate on androgen signaling in castrate-resistant prostate cancer in bone. J Clin Oncol. 2012;30:637-643 10.1200/JCO.2010.33.7675 Crossref
  • [10] E.S. Antonarakis, C. Lu, H. Wang, B. Luber, M. Nakazawa, J.C. Roeser, et al. AR-V7 and resistance to enzalutamide and abiraterone in prostate cancer. N Engl J Med. 2014;371:1028-1038 10.1056/NEJMoa1315815 Crossref
  • [11] E.J. Small, J. Huang, J. Youngren, A. Sokolov, R.R. Aggarwal, G. Thomas, et al. Characterization of neuroendocrine prostate cancer (NEPC) in patients with metastatic castration resistant prostate cancer (mCRPC) resistant to abiraterone (Abi) or enzalutamide (Enz): preliminary results from the SU2C/PCF/AACR west coast prostate cancer. J Clin Oncol. 2015;33 (suppl; abstr 5003)
  • [12] Y. Shiozawa, J. Takita, M. Kato, M. Sotomatsu, K. Koh, K. Ida, et al. Prognostic significance of leukopenia in childhood acute lymphoblastic leukemia. Oncol Lett. 2014;7:1169-1174 10.3892/ol.2014.1822
  • [13] Z. Su, Y.-P. Mao, P.-Y. OuYang, J. Tang, X.-W. Lan, F.-Y. Xie. Leucopenia and treatment efficacy in advanced nasopharyngeal carcinoma. BMC Cancer. 2015;15:429 10.1186/s12885-015-1442-3 Crossref
  • [14] O. Brosteanu, D. Hasenclever, M. Loeffler, V. Diehl. Low acute hematological toxicity during chemotherapy predicts reduced disease control in advanced Hodgkin’s disease. Ann Hematol. 2004;83:176-182 10.1007/s00277-003-0727-9 Crossref
  • [15] Y.H. Kim, H.H. Chung, J.W. Kim, N.-H. Park, Y.-S. Song, S.-B. Kang. Prognostic significance of neutropenia during adjuvant concurrent chemoradiotherapy in early cervical cancer. J Gynecol Oncol. 2009;20:146-150 10.3802/jgo.2009.20.3.146 Crossref
  • [16] K. Shitara, K. Matsuo, D. Takahari, T. Yokota, T. Shibata, T. Ura, et al. Neutropenia as a prognostic factor in advanced gastric cancer patients undergoing second-line chemotherapy with weekly paclitaxel. Ann Oncol. 2010;21:2403-2409 10.1093/annonc/mdq248 Crossref
  • [17] M. Di Maio, C. Gridelli, C. Gallo, F. Shepherd, F.V. Piantedosi, S. Cigolari, et al. Chemotherapy-induced neutropenia and treatment efficacy in advanced non-small-cell lung cancer: a pooled analysis of three randomised trials. Lancet Oncol. 2005;6:669-677 10.1016/S1470-2045(05)70255-2 Crossref
  • [18] K. Shitara, K. Matsuo, D. Takahari, T. Yokota, Y. Inaba, H. Yamaura, et al. Neutropaenia as a prognostic factor in metastatic colorectal cancer patients undergoing chemotherapy with first-line FOLFOX. Eur J Cancer. 2009;45:1757-1763 10.1016/j.ejca.2009.01.019 Crossref
  • [19] Y. Han, Z. Yu, S. Wen, B. Zhang, X. Cao, X. Wang. Prognostic value of chemotherapy-induced neutropenia in early-stage breast cancer. Breast Cancer Res Treat. 2012;131:483-490 10.1007/s10549-011-1799-1 Crossref
  • [20] L. Rambach, A. Bertaut, J. Vincent, V. Lorgis, S. Ladoire, F. Ghiringhelli. Prognostic value of chemotherapy-induced hematological toxicity in metastatic colorectal cancer patients. World J Gastroenterol. 2014;20:1565-1573 10.3748/wjg.v20.i6.1565 Crossref
  • [21] G.R. Pond, W.R. Berry, M.D. Galsky, B.A. Wood, L. Leopold, G. Sonpavde. Neutropenia as a potential pharmacodynamic marker for docetaxel-based chemotherapy in men with metastatic castration-resistant prostate cancer. Clin Genitourin Cancer. 2012;10:239-245 10.1016/j.clgc.2012.06.004 Crossref
  • [22] D. Lorente, J. Mateo, A.J. Templeton, Z. Zafeiriou, D. Bianchini, R. Ferraldeschi, et al. Baseline neutrophil-lymphocyte ratio (NLR) is associated with survival and response to treatment with second-line chemotherapy for advanced prostate cancer independent of baseline steroid use. Ann Oncol. 2014;:750-755 10.1093/annonc/mdu587
  • [23] R.J. van Soest, J. Templeton a, F.E. Vera-Badillo, F. Mercier, G. Sonpavde, E. Amir, et al. Neutrophil-to-lymphocyte ratio as a prognostic biomarker for men with metastatic castration-resistant prostate cancer receiving first-line chemotherapy: data from two randomized phase III trials. Ann Oncol. 2014;2014:743-749 10.1093/annonc/mdu569
  • [24] A.J. Templeton, C. Pezaro, A. Omlin, M.G. McNamara, R. Leibowitz-Amit, F.E. Vera-Badillo, et al. Simple prognostic score for metastatic castration-resistant prostate cancer with incorporation of neutrophil-to-lymphocyte ratio. Cancer. 2014;:2013 10.1002/cncr.28890
  • [25] A.J. Templeton, M.G. McNamara, B. Šeruga, F.E. Vera-Badillo, P. Aneja, A. Ocaña, et al. Prognostic role of neutrophil-to-lymphocyte ratio in solid tumors: a systematic review and meta-analysis. J Natl Cancer Inst. 2014;106:dju124 10.1093/jnci/dju124 Crossref
  • [26] H.S. Han, L.A. Rybicki, K. Thiel, M.E. Kalaycio, R. Sobecks, A. Advani, et al. White blood cell count nadir following remission induction chemotherapy is predictive of outcome in older adults with acute myeloid leukemia. Leuk Lymphoma. 2007;48:1561-1568 10.1080/10428190701474373 Crossref
  • [27] I.F. Tannock, R. de Wit, W.R. Berry, J. Horti, A. Pluzanska, K.N. Chi, et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med. 2004;351:1502-1512 10.1056/NEJMoa040720 Crossref
  • [28] R.M. Franke, M.A. Carducci, M.A. Rudek, S.D. Baker, A. Sparreboom. Castration-dependent pharmacokinetics of docetaxel in patients with prostate cancer. J Clin Oncol. 2010;28:4562-4567 10.1200/JCO.2010.30.7025 Crossref
  • [29] C.J. Sweeney, Y.-H. Chen, M. Carducci, G. Liu, D.F. Jarrard, M. Eisenberger, et al. Chemohormonal therapy in metastatic hormone-sensitive prostate cancer. N Engl J Med. 2015; 10.1056/NEJMoa1503747 150805140037008
  • [30] James ND, Sydes MR, Mason MD, Clarke NW, Dearnaley DP, Spears MR, et al. Docetaxel and/or zoledronic acid for hormone-naïve prostate cancer: first overall survival results from STAMPEDE (NCT00268476) | 2015 ASCO Annual Meeting | Abstracts | Meeting Library n.d. http://meetinglibrary.asco.org/content/147721-156 (accessed August 22, 2015).
  • [31] G. Gravis, J.-M. Boher, F. Joly, M. Soulié, L. Albiges, F. Priou, et al. Androgen deprivation therapy (ADT) plus docetaxel versus ADT alone in metastatic non castrate prostate cancer: impact of metastatic burden and long-term survival analysis of the randomized phase 3 GETUG-AFU15 trial. Eur Urol. 2015; 10.1016/j.eururo.2015.11.005
  • [32] A. Heidenreich, S. Bracarda, M. Mason, H. Ozen, L. Sengelov, I. Van Oort, et al. Safety of cabazitaxel in senior adults with metastatic castration-resistant prostate cancer: results of the European compassionate-use programme. Eur J Cancer. 2014;50:1090-1099 10.1016/j.ejca.2014.01.006 Crossref

Footnotes

a Department of Oncology, University Hospital Zurich, Rämistrasse 100, 8091 Zurich, Switzerland

b Department of Internal Medicine, Stadtspital Waid, Tièchestrasse 99, 8037 Zurich, Switzerland

c Clinical Trial Unit, CTU, University Hospital of Basel, Schanzenstrasse 55, 4031 Basel, Switzerland

d Erasmus MC, and ErasmusMC Cancer Institute, PO Box 5201, 3008 AE Rotterdam, The Netherlands

e Royal Marsden NHS Foundation Trust and the Institute of Cancer Research, Downs Road, Sutton, Surrey SM2 5, UK

f Department of Medicine & Urology, Tulane Cancer Center, 1430 Tulane Avenue, SL-42, New Orleans, 70112 LA, USA

g Department of Oncology, University Hospital of Basel, Petersgraben 4, 4031 Basel, Switzerland

Corresponding author:

1 SvF and DV contributed equally to this study.

This study was funded by Sanofi.