preheader BJMO

Treatment breaks acceptable in standard first-line therapy of advanced renal cell carcinoma

The STAR trial was designed to determine if a tyrosine kinase inhibitor drug-free interval strategy (DFIS) was non-inferior to a conventional continuation strategy (CCS) in the first-line treatment of advanced renal cell carcinoma. Although the results on overall survival just fell short of the non-inferiority boundary, non-inferiority was shown for quality adjusted life years. Based on these results, the investigators concluded that a DFIS is acceptable in this setting.

As treatment with tyrosine kinase inhibitors (TKIs) is associated with significant toxicities and expense, there is increasing interest in using treatment breaks, to reduce toxicity and save money for health services, without compromising efficacy. Therefore, the STAR trial opened in 2012, after sunitinib was the first TKI to be approved for the first-line treatment of renal cell carcinoma (RCC) in the United Kingdom.

STAR trial design

STAR is a phase II/III multicentre, randomised controlled trial conducted in the United Kingdom. Patients with newly diagnosed clear cell metastatic RCC were randomised (1:1) to the drug-free interval strategy (DFIS) or conventional continuation strategy (CCS). After 24 weeks of sunitinib/pazopanib treatment, DFIS patients took a treatment break, until disease progression, with additional breaks dependent on disease response and patient/clinician choice. Trial strategy continued until intolerance, progression on treatment or death. Both co-primary endpoints of overall survival (OS) and quality adjusted life years (QALYs) must demonstrate a pre-defined non-inferiority (≤ 7.5% difference in OS; ≤ 10% difference in QALYs) in intention-to-treat (ITT) and per-protocol (PP) analyses for non-inferiority to be concluded.


In total, 920 patients were randomised (461 to CCS and 459 to DFIS) from January 2012 to September 2017. Median follow-up for the ITT population was 58 months and 78.6% of the QoL data booklets were returned. Baseline patient characteristics were generally well balanced between study arms with 75% of patients having a previous nephrectomy. Median patient’s age (66 years) and proportion of patients with an ECOG performance status of 1 (44.7%) reflect the pragmatic design, as near to ‘real-world’ patients in the United Kingdom as possible. In the DFIS arm, 56.3% of patients continued on trial post-week 24. Overall, a similar number of treatment cycles were received in each arm (median of 5 in CCS and 4 in DFIS). In the DFIS arm, the median length of treatment break was 87 days and 42.7% of patients had two or more (up to nine) treatment breaks.

For non-inferiority in OS to be concluded, the lower bound of the 95% confidence interval for the treatment effect from an adjusted Cox proportional hazards model must be above 0.812 in both the ITT and PP analysis. This was not the case.  In the ITT population HR[95%CI] was 0.97[0.83-1.12] and for PP 0.94[0.80-1.09]. This difference in conclusion in the OS analysis thus precludes confirmation of non-inferiority. The analysis did have reduced power due to the trial stopping, before the required number of events was reached for 80% power. However, consistent non-inferiority was found for QALYs in both the ITT (treatment effect of -0.05) and PP (treatment effect of 0.04) populations. In addition, at two years, DFIS was associated with cost savings of £6,954 per patient due to reduced treatment costs. Both time to strategy failure and summative progression-free interval showed significant differences in favour of the DFIS arm when considered in an adjusted Cox proportional hazard model. On consideration of serious adverse events, deemed to be related to TKI treatment, a smaller proportion of participants in the DFIS arm experienced an event (9.4% vs. 11.7%) and participants in the DFIS arm accounted for fewer of the overall events compared to the CCS arm (39% vs. 61%).


Treatment breaks were acceptable to patients and clinicians, were not detrimental to overall survival or quality of life and had significant cost savings. Overall, 42.7% of patients in the DFIS arm who continued post week 24 had multiple treatment breaks. Further exploratory analysis will consider the characteristics of patients who benefitted from a treatment break. Although immunotherapy is now first-line therapy for many patients, TKIs remain the most appropriate treatment for some patients in the first-line setting, and many others in the second-line setting.


Brown JE, Royle KL, Ralph C, et al. STAR. A Randomised Multi-Stage Phase II/III Trial of Standard first-line therapy (sunitinib or pazopanib) Comparing Temporary Cessation with Allowing Continuation, in the treatment of locally advanced and/or metastatic Renal Cancer (RCC). Presented at ESMO 2021; Abstract LBA28.

Speaker Janet Brown

Janet Brown

Janet Brown, MD, PhD, University of Sheffield, Leeds, UK


See: Keyslides

Back to Top