Activity and Safety of Sunitinib in Poor Risk Metastatic Renal Cell Carcinoma Patients

ARTICLE INFO


INTRODUCTION
Over the past few years, several new agents have been granted approval for the first-line treatment of metastatic renal cell carcinoma (mRCC).These agents include inhibitors of the vascular endothelial growth factor (VEGF) pathway (sunitinib, pazopanib, bevacizumab) and inhibitors of the mammalian target of rapamycin (mTOR) pathway (temsirolimus) (1).
With the exception of the trial investigating temsirolimus (2), pivotal large randomized phase III clinical trials supporting the use of these medications in mRCC excluded or underrepresented patients with poor risk features.However, the efficacy and safety of targeted therapies such as sunitinib in this population is less clear (3).Recently published data suggests that at least 30% of mRCC patients receiving VEGF pathway inhibitors belong to the poor-risk group based on the International Database Consortium (IDC) prognostic model (4).Therefore, we performed a retrospective analysis to evaluate the efficacy, safety and treatment patterns of sunitinib in a non-selected population of poor-risk mRCC patients.

Study design and population
This study was a retrospective analysis of metastatic RCC patients' electronic charts.All patients were treated with sunitinib at two major cancer centers located in the city of São Paulo, Brazil (Instituto do Câncer do Estado de São Paulo, Universidade de São Paulo and Centro de Oncologia, Hospital Sirio-Libanês), between October 2009 and July 2013.Patients were identified from the hospitals' administrative databases, and relevant information was retrieved from electronic medical records.This analysis was approved by the Institutional Review Boards of both institutions.
The primary study objective was to determine the population's overall survival (OS).Secondary objectives were the following: examination of the safety profiles, frequency of treatment modifications, time to treatment discontinuation and response rate in an outside clinical trial setting.
The patient inclusion criteria were (a) diagnosis of metastatic RCC of any histologic type; (b) poor risk features according to the IDC model ( 4); (c) a Karnofsky performance status (KPS) of 60 or more at baseline assessment and (d) anti-VEGF--naïve patients before starting sunitinib.

Assessment
Trained physicians extracted patients treatment data from electronic medical records, including the following information: date of RCC diagnosis, demographic variables, comorbidities, metastatic site(s), baseline KPS, drug-related AE data, laboratory data, and the results of key radiological tests.Additionally, the first and last dates of sunitinib use, treatment modifications, and both baseline and follow-up tumor measurements were also recorded.A patient's observation period began on the date of initiation of sunitinib and ended either at the time of their last center visit or death.

Safety
Safety outcomes included the numbers and proportion of patients who experienced spe-cific adverse events (AE), of any grade and of grade 3 or higher.We retrospectively assessed AE and assigned grade levels based on the National Cancer Institute's Common Terminology Criteria for Adverse Events (CTCAE), version 3.0, at chart abstraction because grade levels of adverse events are not regularly recorded in medical charts in clinical settings (5).

Treatment Patterns
Treatment patterns included the numbers and proportion of patients who discontinued treatment or had dose modifications during sunitinib treatment and those who switched to a second-line treatment.The type, date, and the reasons for treatment modification were obtained from the patients' medical charts.Time to treatment discontinuation (TTD) was defined as the time from starting sunitinib to the date of the last administered dose independent of cause for discontinuation or loss to follow-up, whichever occurred first.

Efficacy
Overall survival was defined as the interval between the start of therapy and death from any cause (with survival times censored at last follow-up for patients alive at the time of last assessment).The response assessment was reviewed by an independent radiologist, using guidelines for change in the sum of maximal diameters as defined by Response Evaluation Criteria In Solid Tumors, but responses were not confirmed by a second assessment (6).

Statistical analysis
The patient baseline characteristics, overall response rate (ORR), AEs and sunitinib treatment patterns were reported based on descriptive statistics.Means, medians and ranges were used to describe continuous variables, and frequencies or percentages were used for categorical variables.TTD and OS were estimated by the Kaplan-Meier method, and their 95% confidence interval (CI) were calculated.All data analyses were conducted using SPSS, version 21.0, Armonk, NY.

Treatment patterns
Table-2 summarizes the sunitinib treatment patterns and reasons for discontinuation.Thirty seven patients (69.8%) started on therapy at the standard schedule of 50mg once daily, 4 weeks on, followed by 2 weeks off; sixteen patients (30.2%) started treatment at a reduced dose.Although not statistically significant, patients on standard regimen were younger (median age: 58.6 versus 68.8 years; p=0.1) than those on reduced dose regimens.Median TTD was 3.30 months (95% CI: 1.96-4.63).At the time of analysis, almost all patients had discontinued therapy (96.2%), mostly due to disease progression (69.8%).Drug toxicity caused treatment interruption in 26.4% of patients.Furthermore, adverse events required dose modifications in 26.4% of patients.Three patients (5.7%) experienced dose escalation after starting treatment at a reduced dose.

Safety
Adverse events reported in patients' medical charts are summarized in Table-3.Among patients included in this retrospective analysis, 94.3% experienced at least one AE, including grade 3 or higher AE.The most common all-grade AE was fatigue or asthenia (64.1%), followed by nausea (49.0%), stomatitis (41.5%), vomiting (37.7%) and hypothyroidism (32.1%).Grade 3 or higher AE was reported in 39.6% of patients.The most common grade 3 or higher AE were fatigue (15.1%), neutropenia (9.5%), nausea, vomiting, and diarrhea (7.5% each).There was one treatment-related death due to febrile neutropenia and septic shock.There was no statistically significant difference in the frequency of grade 3 or higher AEs between patients who started sunitinib at the standard dose and patients who initiated therapy at a reduced dose regimen.

Efficacy
Of the 53 patients included in this analysis, nine (17.0%) achieved partial response, 12 (22.6%)had stable disease and 13 (24.5%)had progressive disease.Nineteen patients (35.8%) were not evaluated for response, including 14 patients (26.4%) with early discontinuation: one treatment-related death caused by febrile neutropenia; one loss to follow-up before response evaluation; and 4 (7.5%)cases of early discontinuation due to limiting AE.No complete responses were seen.At a median follow-up of 7.5 months, 9 patients (16.9%) were still alive.The estimated median OS was 8.16 months (95% CI, 5.73-10.59)(Figure -1).There was no statistically significant difference in median OS between patients who started sunitinib at the standard dose (9.1 months; 95% CI, 6.12-12.07)and patients with modified regimens (7.2 months; 95% CI, 6.28-8.11).Only 6 patients received a subsequent treatment after discontinuing sunitinib (five of them were treated with interferon-alpha and one was treated with everolimus).

DISCUSSION
Data regarding the use of sunitinib in patients with poor-risk mRCC are scarce and therefore, data on efficacy and safety of this therapy in this population are less known.Although patients experienced an OS of 8.16 months in this analysis, a high proportion of grade 3 or higher AEs occurred (39.6%) and 26.4% of patients discontinued treatment due to toxicity, regardless of starting treatment at a reduced dose.
Here, we describe a median OS of 8.16 months (95% CI, 5.73-10.59),with only 11.3% of patients exposed to second line therapy, which hi-ghlights a potential survival benefit of sunitinib in the setting of poor-risk patients.This finding is consistent with the results from both the IDC original and external validation datasets demonstrated that poor-risk patients show a median OS of 8.8 and 7.8 months, respectively (4, 7) and compares favorably to subset analysis of phase III and IV trials (8,9).In the pivotal phase III trial only 6.4% patients with poor-risk MSKCC criteria were included, rendering a median OS of 5.3 months (95% CI: 4.2-10.0months) with sunitinib compared with 4.0 months (95% CI: 2.7-7.2months) with IFN-alpha (HR: 0.660; 95% CI: 0.360-1.207)(8).Similarly, the expanded access program (EAP) included 9% poor-risk patients according MSKCC criteria (373 out of 4564 total subjects) and found the same median OS of 5.3 months (95% CI: 4.6-5.4months) (9).It must be highlighted that the numerical variations in median OS described across these different studies might be, at least in part, due to the fact of different poor risk classification, since 14% of MSKCC intermediate-risk patients are reclassified as poor-risk when stratified by IDC criteria (4).
In our series, an ORR of 17.0% was achieved and 22.6% of the evaluable patients achieved stable disease.No literature data on response rate restricted to poor risk patients is available; however, in EAP trial, which included different risk subgroups, an ORR of 17%, was reported (9).Considering that 26.4% of our population died before response assessment, it is possible that the rate of patients who truly benefit from sunitinib is really lower.However, the lack of a predefined interval for response assessment in our retrospective series could also justify, at least in part, the lower ORR of single-agent reported here.
Other main issues addressed by our study were safety and treatment patterns.Although grade 1 and 2 AE might have been underreported, a high proportion (39.6%) of patients experienced grade 3 or higher AEs.In our series, the rates of grade ≥3 fatigue, nausea, vomiting, and diarrhea are higher than those previously reported in both controlled and non-controlled studies (Table-4) (8)(9)(10)(11)(12).The high proportion of serious AE might be explained by the broader eligibility criteria.Some of the included patients had poor KPS (60-70), which are commonly treated under routine care, albeit excluded from randomized clinical trials (13).Furthermore, our analysis showed that 26.4% of patients discontinued sunitinib due to AE, which is higher than rates of 8-20% in other studies (9,10,14).In this study, 30.2% of our patients started at the standard dose of standard dose of sunitinib.This finding might reflect the concern of AE on patients with poor KPS and older age treated with sunitinib.A recent analysis showed that approximately 30% of elderly patients treated with sunitinib start therapy with reduced dose (15).However, in our analysis there were no were statistically significant difference in rates AE between these two groups of patients.
Although previous studies have examined activity and safety of sunitinib outside a clinical trial, to our knowledge, this is the first observational study that specifically addressed the efficacy, safety and treatment patterns of this treatment in a poor-risk population according IDC model.Furthermore, our data reflect a "real-world" practice setting, and OS could be accurately estimated and was consistent with that reported in the literature.

CONCLUSIONS
Our study confirms that sunitinib can be a feasible and active treatment for some patients with mRCC and poor risk features.However, the high incidence of AE and the drug discontinuation rates suggest a need for careful monitoring.Considering the incurable nature of advanced RCC, this retrospective study indicates that continued research efforts to identify more effective and better tolera-ted treatments for mRCC are needed, especially for poor risk patients.

Table 3 -Reported adverse events during sunitinib treatment*.
*Adverse events experienced by at least 10% of patients are reported; +Including a grade 5 adverse event.

Table 4 -Reported grade 3 or higher adverse events in studies evaluating sunitinib in mRCC population.
mRCC: metastatic renal cell carcinoma; *current paper