Abstract
Background/Aim: Sunitinib is the current standard of care for first-line (1L) treatment of metastatic renal cell carcinoma (mRCC). Previous studies suggest that a modified treatment schedule may benefit patients. Our aim was to evaluate efficacy and safety regarding sunitinib treatment modification in 1L treatment of mRCC. Materials and Methods: Data were drawn from STAR-TOR, a German real-world registry to evaluate outcomes of patients with mRCC who received 1L sunitinib. Patients were divided into two groups: subsequent treatment modification (SM) or remaining on standard dose/schedule (SS). Time on treatment (TT), progression-free survival (PFS), and overall survival (OS) were estimated. Results: Overall, 297 patients were analyzed; 33% underwent treatment modification. Significant baseline differences between groups were observed; SM patients were older and had a more favourable Karnofsky performance status. SM patients achieved better outcomes than SS patients for median TT (15.1 versus 3.9 months; p<0.0001), PFS (15.1 versus 6.0; p<0.0001), and OS (38.1 versus 13.7; p<0.0001). Diarrhoea (34%/17%), fatigue (30%/11%), hand-foot syndrome (28%/10%), and stomatitis (20%/6%) were more frequently reported in SM versus SS; incidence was reduced following schedule/dose modification (except diarrhoea). Conclusion: In addition to AE mitigation, sunitinib treatment modification may help improve efficacy outcomes in mRCC by prolonging treatment duration.
Sunitinib is an oral, multitargeted tyrosine kinase inhibitor with antitumor activity towards various neoplasms (1). Sunitinib is approved for the treatment of metastatic renal cell carcinoma (mRCC) (2). It is one of the most studied targeted therapies for this indication, with proven efficacy and a manageable safety profile, supported by robust clinical trial data as well as extensive use in clinical practice for over 10 years (2). Therefore, it represents a current standard of care for first-line treatment of mRCC (2).
The standard dosing schedule for sunitinib is 50 mg/day for a 4-week period followed by 2 weeks off – a 4/2 schedule (3). It is well established that increased systemic exposure to sunitinib is associated with improved patient outcomes, including longer progression-free survival (PFS), longer overall survival (OS), and greater tumour response rates, with the standard dose (50 mg/day) offering a reasonable balance between clinical benefit and risk of adverse effects (4). However, on average 50% of patients receiving sunitinib treatment at the standard dose and schedule require dose reductions to mitigate toxicity, which lowers both dose intensity and drug exposure, which in turn may compromise efficacy (5, 6). This consideration necessitated the exploration of modified dose schedules in a real-world setting to improve tolerability without compromising efficacy and to establish the optimal risk:benefit ratio.
Modified treatment schedules of sunitinib, such as using a different number of days on/off therapy from the standard schedule based on individual safety and tolerability, e.g. 2 weeks on/1 week off (2/1), and dose-individualization strategies, have been shown to further improve tolerability and quality of life in patients receiving sunitinib therapy (7-16). Patients who switch to a modified treatment schedule, e.g. 2/1, exhibit significantly fewer adverse events (AEs) of all grades (e.g. diarrhoea, hand-foot syndrome [HFS], hypertension, and fatigue), including grade 3/4 AEs compared with those remaining on the standard schedule (15, 16). Fatigue and HFS, which are common toxicities with the 4/2 sunitinib schedule, were significantly less frequent when sunitinib was given as the 2/1 schedule (p=0.0003 and p=0.0004, respectively) (10, 15). Moreover, Miyake et al. reported that the sunitinib 2/1 schedule was associated with better health-related quality of life, as evidenced by more favourable scores in all eight sub-scores of the Medical Outcomes Study Short-Form (36-item) Health Survey compared with the 4/2 schedule (16). Dose individualization provides another mechanism to reduce toxicity with sunitinib treatment (17, 18).
Treatment modification was also associated with a significantly prolonged time on treatment (TT) according to previous studies (9, 15). However, existing data are sparse or inconclusive regarding PFS and OS in patients undergoing treatment modifications (modified schedule and/or dose) compared with those who remain on the standard treatment regimen because of methodological limitations, e.g. single-centre experiences with limited numbers of patients and retrospective designs (13, 15). Furthermore, it is currently unclear which observable factors (including toxicities and performance status) influence a physician's decision to modify the schedule and/or the dose of sunitinib in clinical practice.
The objective of this study was to evaluate efficacy and safety outcomes in relation to sunitinib treatment modification, defined as an alternative treatment interval (on therapy/off therapy) or dose, in a cohort of German patients with mRCC receiving first-line sunitinib. A further objective was to identify the factors associated with sunitinib treatment modification.
This analysis will provide valuable insights into the current use of sunitinib schedule/dose modifications in a German real-world setting and will facilitate a more evidence-based consideration of the role of modified sunitinib schedules/doses in the management of patients with mRCC.
Materials and Methods
Data source. Anonymous data up to February 2017 were extracted from the STAR-TOR study – a large prospective German multicentre registry for the evaluation of the safety, tolerability, and efficacy of temsirolimus, sunitinib, and axitinib in the treatment of patients with mRCC, mantle cell lymphoma, and gastrointestinal stroma tumour (NCT00700258). Details regarding the registry design and data collection have been previously described elsewhere (19-21).
Data were retrospectively analysed from patients with mRCC who were initiated on the standard sunitinib schedule of 50 mg/day 4/2 as first-line treatment. The following clinical data were recorded in the registry database and extracted for analysis: nephrectomy status, histology, International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) prognostic criteria [Karnofsky performance status (KPS), anaemia, neutrophilia, thrombocytosis, hypercalcaemia, and time from diagnosis to treatment (22)], comorbidities, sites of metastases, sunitinib doses/schedules throughout the treatment period, assessment of response to therapy according to Response Evaluation Criteria in Solid Tumors (RECIST), treatment-related toxicities, and mortality.
Patient groups. Patients were categorized into two treatment groups: (1) sunitinib initiated at standard dose/schedule with subsequent dose/schedule modification (SM); (2) sunitinib at standard schedule/dose only without subsequent dose/schedule modification (SS).
Patients were also categorized according to their IMDC prognostic score at enrolment as favourable (0 risk factors), intermediate (1/2 risk factors), or poor (3+ risk factors) risk.
Main outcomes. TT, PFS, and OS were analysed. TT was defined as the time from initiation of sunitinib to discontinuation or death from any cause. PFS was defined as the time from initiation of sunitinib to documented disease progression (assessed by RECIST) or death from any cause. OS was defined as the time from initiation of sunitinib to death from any cause.
The incidences of treatment-related AEs (TRAEs) and severe TRAEs (grade ≥3) during treatment with sunitinib were estimated in both treatment groups. For SM patients, the change in incidence of TRAEs between the two treatment periods (before versus after dose/schedule change) was evaluated.
Statistical analyses. Descriptive statistics were used to summarize the data.
Median TT, PFS, and OS, with associated 95% confidence interval (CIs), were estimated using the Kaplan–Meier method. Differences between treatment groups were assessed using log-rank tests. Stratified log-rank tests by IMDC risk group were also performed to assess differences at baseline.
A multiple logistic regression model was constructed to identify the determinants associated with dose/schedule modifications. Covariates included: demographics, nephrectomy status, histology, KPS, and specific TRAEs that are commonly reported with sunitinib. Adjusted odds ratios (ORs) and CIs are reported for each covariate.
All analyses were performed in Stata version 15.0 or later (StataCorp LLC, College Station, TX, USA) software.
Results
Patient characteristics. In total, 297 patients were treated with first-line sunitinib initiated at the standard dose/schedule between June 2010 and January 2017 and were included in the analysis (Figure 1). One-third (33%) of patients had a treatment modification (SM group).
Treatment modifications mainly included dose reductions using 37.5 mg 4/2 (n=37) and 25 mg 4/2 (n=29) or changing to a 2/1 schedule with the 50 mg dose (n=16) (Table I).
Patients' baseline characteristics are reported in Table II. Differences observed were: patients in the SM group were older (median age 69 years versus 65 years for SS group; p=0.0230) and more of them had a favourable KPS at enrolment compared with the SS group (88.4% versus 78.2%, respectively; p=0.0330). A similar pattern was seen for IMDC prognostic score, with significantly more patients in the SM versus SS group being classified as having a favourable or intermediate risk (p<0.05). Details of patients' IMDC risk factors at baseline are given in Table III.
CONSORT diagram. 4/2, 4 weeks on/1 week off; mRCC, metastatic renal cell carcinoma.
Time on treatment. The median (95%CI) TT was 15.1 (11.7-18.7) months in the SM group versus 3.9 (3.4-4.6) months in the SS group (p<0.0001) (Figure 2). Significant differences were also observed when stratifying by IMDC risk status at enrolment: in the SM and SS groups, median TT was 13.7 and 3.6 months (p<0.0001), respectively, for patients with an intermediate/poor IMDC prognostic score. In addition to the Kaplan–Meier analysis, descriptive analysis showed that for patients in the SM group, the median (interquartile range) TT with the standard dose/schedule was 3.7 (1.5-6.2) months before treatment modification versus 6.8 (2.8-12.4) months afterwards.
Progression-free survival. Median (95%CI) PFS was 15.1 (11.9-19.2) months in the SM group and 6.0 (4.7-7.6) months in the SS group (p<0.0001) (Figure 3a). In patients with intermediate/poor IMDC prognostic score, median (95%CI) PFS was 14.0 (10.9-19.4) months in the SM group and 4.2 (3.4-5.5) months in the SS group (p<0.0001).
Overall survival. Median (95%CI) OS was significantly improved in SM versus SS patients [38.1 (28.9-50.5) versus 13.7 (10.1-20.2) months, respectively; p<0.0001] (Figure 3b). In patients with intermediate/poor IMDC prognostic score, median (95%CI) OS was 35.6 (19.6-not reached) months and 9.7 (8.5-15.9) months in the SM and SS groups, respectively (p<0.0001).
Modifications of sunitinib dose/schedule from standard 50 mg 4/2 (n=98).
Treatment-related adverse events. TRAEs were reported in 66.7% of patients, of whom 23.6% experienced grade ≥3 events. The most common TRAEs were diarrhoea (22.6%), fatigue (17.2%), taste disturbance (16.8%), HFS (15.8%), thrombocytopenia (12.1%), mucositis (11.1%), nausea (10.8%), stomatitis (10.4%), and leukopenia (10.1%) (Table IV).
TRAEs were more frequently reported in SM versus SS patients, e.g. diarrhoea (SM/SS: 33.7%/17.1%), fatigue (29.6%/11.1%), HFS (27.6%/10.1%), and stomatitis (20.4%/5.5%).
In the SM group, the incidence of any TRAE was significantly reduced after treatment modification compared with the initial standard dose/schedule (68.4% versus 49.0%; p=0.001). The greatest reductions were observed for taste disturbance (before/after modification: 20.4%/4.1%), fatigue (17.3%/3.1%), HFS (18.4%/7.1%), thrombocytopenia (10.2%/3.1%), and mucositis (12.2%/5.1%) (Table V).
The incidence of severe TRAEs was not significantly different after treatment modification compared with the initial standard dose/schedule (14.3% versus 20.4% with maximum toxicity grade ≥3; p=0.24) (Table V). Overall, the incidence of severe TRAEs was significantly higher in the SM versus SS group (34.7% versus 18.1%, respectively; p=0.002).
Determinants of sunitinib treatment modification. Variables associated with increased odds of undergoing a treatment modification were: HFS [OR 95%CI: 2.943 (1.333-6.495)], fatigue [2.688 (1.221-5.917)], diarrhoea [2.547 (1.247-5.201)], and age [1.046 (1.010-1.083)] (Table VI). Comorbidities, nephrectomy status, and histology were not found to be significant predictors of treatment modification. Notably in the regression analysis, a favourable KPS at enrolment was no longer a significant discriminant factor between SM and SS.
Patient demographics and clinical characteristics.
Discussion
Although generally associated with well-characterized acceptable and manageable toxicity, sunitinib is nonetheless associated with AEs that could potentially lead to early treatment discontinuation (23). Several alternative doses and individualized time on/off therapy have been proposed to mitigate AEs, and some studies have suggested that efficacy is maintained with these approaches (9, 15). Additionally, European Association of Urology guidelines for renal cell carcinoma recognize that alternate scheduling of sunitinib is being used to manage toxicity; however, they state that robust data to support the use of alternate scheduling of sunitinib are still lacking (24).
A key finding of the current analysis is that patients who underwent treatment modification experienced certain AEs at a higher overall rate than those who stayed on the standard sunitinib dose/schedule. However, patients with a treatment modification had a longer TT than those who did not, which gave more opportunity to observe AEs and may account for this result. In addition, patients who did not require a treatment modification might have had lower exposure to sunitinib and therefore experienced fewer toxicities. Consistent with the known toxicity profile of the drug (3), experiencing specific AEs, e.g. HFS, fatigue, and diarrhoea, was also found to be the main determinant associated with a treatment modification with sunitinib. These specific AEs have also been found to be predictors of sunitinib efficacy in mRCC (25).
Kaplan–Meier survival plot of time to discontinuation of first-line sunitinib treatment. Median time on treatment (95% confidence interval): SS: 3.9 (3.4-4.6) months vs. SM: 15.1 (11.7-18.7) months; p<0.0001. *One patient excluded with time on treatment equal to 0. SM: Sunitinib initiated at standard dose/schedule with subsequent dose/schedule modification; SS: sunitinib as per standard dose/schedule only without any subsequent dose/schedule modification.
Treatment modification occurred early in treatment, with a median time to modification of 3.7 months, and was associated with prolonged treatment duration. In patients who underwent treatment modification, a significant reduction was observed in the incidence of specific AEs after the change in regimen, including fatigue, HFS, mucositis, thrombocytopenia, and stomatitis. Therefore, we can hypothesize that these patients were more likely to stay on treatment with sunitinib following the dose/schedule adjustment, allowing better management of AEs. Most of the AEs that were managed by treatment modifications were of mild/moderate intensity.
Aligning with previous studies (9, 15), treatment modification with sunitinib does not appear to compromise efficacy (PFS and OS). Indeed, it was found that patients who had their sunitinib dose/schedule modified achieved better efficacy outcomes than those who did not. This might be because of the longer treatment duration with modification – and thus greater sunitinib exposure – leading to optimal treatment efficacy. Conversely, we can hypothesize that patients in the standard sunitinib dose/schedule group might have discontinued treatment prematurely because of toxicity rather than disease progression. If these patients had undergone dose/schedule modification they might have experience improved tolerability, and thus potentially prolonged treatment duration.
Results for median PFS (15.1 months) and OS (38.1 months) in patients who had a treatment modification compare favourably with those reported in the literature for patients treated with first-line sunitinib (8.8-10.8 months and 6.1-31.9 months respectively) (26-28). As a complementary finding, efficacy outcomes were worse in patients who maintained their initial standard dose/schedule of sunitinib than in those who had a treatment modification (median PFS and OS of 6.0 versus 15.1 months and 13.7 versus 38.1 months, respectively). These strong differences between the two patient groups further confirm the results of previous studies (9, 15) showing the benefit for efficacy outcomes of treatment modification and may support the hypothesis that development of sunitinib-related toxicities identifies patients more likely to respond to sunitinib treatment (9).
Kaplan–Meier estimates of (a) progression-free survival and (b) overall survival in patients receiving sunitinib according to their dose/schedule (standard and modified). Median progression-free survival (95% confidence interval): SS: 6.0 (4.7-7.6) months vs. SM: 15.1 (11.9-19.2) months; p<0.0001. Median overall survival (95% confidence interval): SS: 13.7 (10.1-20.2) months vs. SM: 38.1 (28.9-50.5) months; p<0.0001. SM: Sunitinib initiated at standard dose/schedule with subsequent dose/schedule modification; SS: sunitinib as per standard dose/schedule only without any subsequent dose/schedule modification.
IMDC risk factors of patients at baseline.
Treatment-related adverse events (TRAEs).
Treatment-related adverse events (TRAEs) before and after treatment modification.
Logistic regression of key patient characteristics on sunitinib dose/schedule modifications.
Strengths of this analysis were the prospective data collection, including details of dose/schedule modifications and comprehensive recording of toxicities, thorough analysis of safety, and inclusion of AEs in a regression model to identify determinants of treatment modification. In addition, this is the first analysis of treatment modification using STAR-TOR data.
Limitations of our study included unknown patient compliance with treatment and the risk of bias inherent to observational research. The two treatment groups were significantly different at enrolment in terms of age and KPS, and more IMDC intermediate-risk patients underwent treatment modification compared with poor-risk patients. However, Kaplan–Meier estimates adjusted for IMDC criteria showed similar results and only patient age was found to be significantly different between treatment groups in the logistic regression model, which was no longer the case for KPS. In addition, it should be noted that current evidence indicates that treatment duration with sunitinib is shorter in elderly patients (29).
The influence of practice-related factors (physician specialty, office environment, etc.), which can drive the selection of treatment modification with sunitinib, could not be explored. Analysis of individual treatment-modification strategies could not be performed owing to small sample sizes.
It should also be acknowledged that the sample size of patients within the STAR-TOR registry at the time of analysis who underwent a treatment modification was small compared with the number of patients remaining on the standard dose and schedule. Further research on a larger cohort of patients with mRCC treated with sunitinib should be conducted to confirm the findings of this analysis. Furthermore, owing to sample size limitations, the majority of these analyses are descriptive; therefore, the survival results should be carefully considered, and further analysis should be performed on a larger sample to understand the impact of key confounding factors on these efficacy outcomes.
Despite these limitations, and in the absence of randomized controlled trial data, these findings are useful for devising improvements for clinical practice. While previous studies have investigated treatment modifications of sunitinib for the treatment of mRCC, this analysis provides a recent assessment of these adjustments and their impact on efficacy outcomes in Germany. In the absence of disease progression, treating physicians should employ all available strategies to ensure patients are deriving maximum benefit from treatment with sunitinib. Therapy management strategies, such as schedule modification and dose reductions, can help reduce AEs including fatigue, HFS, and diarrhoea, while prolonging treatment duration.
Conclusion
In patients treated with sunitinib as first-line therapy for mRCC, use of a modified schedule and/or dose can help reduce the incidence of some AEs, allowing patients to continue treatment for longer, thereby potentially maximizing the efficacy of sunitinib, leading to prolonged PFS and OS compared to patients who do not undergo any treatment modification. More emphasis on implementation of treatment modification for the management of AEs of special interest is warranted in clinical practice.
Acknowledgements
This publication was funded by Pfizer Inc. The STAR-TOR study is funded by Pfizer Pharma GmbH.
Footnotes
This article is freely accessible online.
Conflicts of Interest
Danielle Bargo, Marcus Hubbe, and Despina Thomaidou are employees of Pfizer. Stuart Blackburn, Nawal Bent-Ennakhil, and Robert Wood are employees of Adelphi Real World. Martin Boegemann received travel support from: BMS, Bayer, Novartis, Janssen, and Amgen; fees for an advisory role from: Janssen, Bayer, Astellas, BMS, MSD, Amgen, Pfizer, Ibsen, Sanofi, and Novartis; and honorary fees for talks from: Janssen, Bayer, Stellas, BMS, Pfizer, Amgen, Novartis, and MSD.
- Received October 17, 2018.
- Revision received October 22, 2018.
- Accepted October 23, 2018.
- Copyright© 2018, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
