In 2010, we have the clinical opportunity to choose among several “targeted” agents when treating patients with metastatic renal cell carcinoma (RCC). Although these agents have clearly improved progression-free survival in patients with metastatic disease, we still are awaiting further clinical trial data to know which agent is best for individual patient. These trials include comparisons of sunitinib vs. pazopanib as front line therapy, sunitinib vs. everolimus as front line therapy, sorafenib vs. temsirolimus and axitinib vs. sorafenib as second line therapy, and many other studies of sequencing these agents, combining already approved agents, and evaluating these agents in specific populations such as non-clear cell RCC. The ultimate goal of targeted therapies is to accurately identify patient subsets that are most likely to benefit from a specific therapy. Consequently, much of current research is dedicated to identifying and validating relevant molecular biomarkers, expressed in either normal host issue or the tumor itself, to predict drug activity and patient benefit as well as mechanisms of resistance to our therapies. Many of the meeting’s genitourinary cancer presentations focused on clinical trials examining new approaches to combining promising agents or using biomarkers to enrich for response to these agents in an effort to create more effective treatments for patients with mRCC.
Primo Lara, MD, chair of the educational session, Evolving Standards of Care in Advanced Renal Cell Carcinoma, noted that the management of advanced renal cell cancer (RCC) in 2010 presents opportunities and challenges for the practicing oncologist never imagined just a decade ago. Once considered among the least treatable of advanced malignancies, practitioners in the United States now have seven approved agents to choose from to treat this disease. The rapid development of these agents in the absence of a historical context of effective treatment strategies has raised many more questions than available answers regarding their optimal deployment in kidney cancer, the role of cytoreductive nephrectomy in this new era, and best to personalize therapy for individual patients.
According to Dr. Lara, the first step toward individualizing therapy in RCC is differentiating between prognostic and predictive markers: prognostic markers provide information about disease outcome independent of a received treatment, whereas a predictive marker provides information based on a specific treatment and only predictive markers can be used to indicate which patients should be treated with a particular agent. Dr. Lara stressed that in the currently available RCC treatments; no baseline predictive markers have been definitively identified, let alone prospectively validated.
Several predictive biomarkers are under evaluation currently and require confirmatory studies. For example, we have shown (abstract 4631) that baseline serum LDH was both prognostic and predictive of survival benefits by temsirolimus, an mTOR inhibitor, in a large phase III study of patients with metastatic poor-risk RCC. Others (McDermott et al, abstract 4514) are examining markers such as CAIX expression in clear cell RCC to predict benefit from high dose IL-2. Finally, Tran et al found a correlation between a multiplexed panel of cytokines (IL-6, IL-8, and HGF) with benefit from pazopanib therapy (abstract 4522). As further studies are performed to validate these findings, it is hoped that these biomarkers will be able to guide the busy clinician in selecting patients for appropriate therapies.
An ideal predictive marker is easily reproducible, correlated with a clinical endpoint, has a quick turn-around time and is available prior to initiation of treatment. Naturally, the clinical goal of predictive biomarkers is the ability to select a particular therapy based on the likelihood of benefit to the patient. As an example of a putative biomarker as a predictive factor, Dr. Lara discussed emerging data on clinical predictive biomarker factors for angiogenesis inhibitors. One post-treatment predictive factor is the development of hypertension. In an analysis of trials with the VEGFR-TKI axitinib, patients with RCC who subsequently develop diastolic blood pressure of 90 mmHg or greater appear to have a higher median survival time (30 months) as compared with a diastolic blood pressure of less than 90 mmHg (9.7 months). It is not clear at this time if the development of hypertension is related to higher drug levels and longer exposures to this drug, or if treatment of hypertension could potentially reduce efficacy to this agent. Further work to evaluate hypertension and its treatment as a predictive factor in randomized controlled trials is needed.
Dr. Lara discussed several other putative predictive markers in the array of therapies approved for treatment of mRCC, concluding that untargeted use of targeted therapies characterizes the current state of clinical RCC management. Therefore, identification and validation of predictive markers is among our most critical goals. To that end, examining molecular strategies is encouraged in the next generation of RCC trials. Dr. Lara predicted that true individualized therapy that selects patients with RCC for specific treatments based on the molecular biology of the specific tumor will soon enhance if not replace the current paradigm of selection based on clinical features. In this context, the findings of Rini et al (abstract 4501) looking at a genomic classification of primary RCC patients are noteworthy, as they found distinct populations of clear cell RCC patients with widely disparate prognoses based on molecular signatures of angiogenesis and proliferation, even after adjusting for clinical factors. These data highlight the need for further exploration of biomarkers to help guide prognostication and treatment decisions in RCC.
Results of the randomized TORAVA phase II trial.
In the oral abstract session, B. J. Escudier, MD, presented data from the TORAVA study, which was designed to determine whether the combination of temsirolimus and bevacizumab would improve the outcomes in metastatic renal cell carcinoma (mRCC) by leading to a synergistic or additive effect with an acceptable safety profile in first-line treatment. Since the temsirolimus and bevacizumab combination had proven feasible at full doses of each agent with promising early activity in phase I, Dr. Escudier said that the simple rationale for the study was asking the question, “Are two drugs really better than one,” in this population.
TORAVA was a multicenter, randomized phase II study with three parallel groups: Untreated mRCC patients with ECOG Performance Status (PS) of 2 and measurable disease were randomized to temsirolimus-bevacizumab combination (arm A), sunitinib (arm B) or bevacizumab and ?-interferon (arm C). The patients were treated until disease progression or the development of an unacceptable toxicity.
The primary objective was to estimate the non-progression rate at 48 weeks for arm A. Using Fleming's single-stage procedure and a non-progressive rate (NPR48) at week 48 of 50% as the minimum needed for the combination to warrant further study, the planned sample size for arm A was 80 patients (assuming NPR48 of 35% in the pivotal trials [Motzer et al 2007, Escudier et al 2007]).
Major secondary endpoints were toxicity, response rate and survival. From 2008 to 2009, 171 patients were randomized: 88 (51%), 42 (25%) and 41 (24%) to arms A, B and C, respectively. Treatments were prematurely stopped for other reasons than progression in 43% (A), 12% (B) and 23% (C). Grade 3/4 events were observed in 36%, 14% and 27% pts in arms A, B, C respectively; two toxic deaths occurred in arm A.
From 03-2008 to 05-2009, 171 patients were randomized: 88 (51%), 42 (25%) and 41 (24%) to arms A, B and C, respectively. An intent-to-treat analysis with a median follow-up of 43 weeks was performed. Treatments were prematurely stopped for other reasons than progression in 43% (A), 12% (B) and 23% (C). Grade 3/4 events were observed in 36%, 14% and 27% pts in arms A, B, C respectively; two toxic deaths occurred in arm A. In an intent-to-treat analysis with a median follow-up of 43 weeks, NPRs-48 were 43.2% (95% CI, 32.7-54.2), 47.6% (95% CI, 32.0-63.6) and 65.9% (95% CI, 49.4-79.9) in arms A, B and C respectively. Best response rates (RECIST) were 25%, 24% and 34% respectively.
The toxicity profile of the combination was higher than expected, leading to a high drop-out rate. Dr. Escudier concluded that the results do not suggest any evidence of a synergistic/additive efficacy of this combination. Patients will continue to be followed and updated data will be reported. These data suggest that more is not necessarily better with combination mTOR/VEGF inhibition, and toxicity can certainly limit the safety of this approach. Combined with the significant toxicity seen with temsirolimus-sunitinib combinations, temsirolimus-everolimus combinations, and bevacizumab-sunitinib combinations, the use of these combinations and others should be limited to the clinical trial setting.
In a poster session, titled, “A Phase II Study Testing the Safety and Activity of AGS-003 as an Immunotherapeutic in Subjects with Newly Diagnosed Advanced Stage Renal Cell Carcinoma (RCC) in Combination with Sunitinib,” lead author, Asim Amin, MD, PhD, discussed data from a trial that evaluated the clinical activity, safety, and immune response of AGS-003 treatment, given in combination with sunitinib, in patients with newly diagnosed advanced renal cell carcinoma (RCC). ABS-003 is an autologous (self) tumor vaccine composed of dendritic cells pulsed with an individual’s own cancer cells and CD40, a co-stimulatory immune molecule.
The open-label phase II trial enrolled 22 evaluable, newly diagnosed post-nephrectomy patients with metastatic clear-cell RCC. The patients were first received 4 weeks of sunitinib, followed by 2 weeks off treatment; patients then received five intra-dermal injections of AGS-003 every 3 weeks in combination with sunitinib, followed by every 3 months until disease progression or study end. AGS-003 was well tolerated, with no immunotherapy-related serious adverse events or grade 3/4 adverse events reported. Correlative immune response evaluation is ongoing.
According to Dr. Amin, the combination of AGS-003 plus sunitinib in this intermediate to poor risk prognosis group of advanced RCC patients has been encouraging, with a PFS of 12.5 months. Interim results demonstrate that 81% of patients experienced clinical benefit, as defined by partial response or stable disease as per RECIST. Based on early results, the combination of a personalized immunotherapy, such as AGS-003, with a standard treatment like sunitinib, represents a potentially promising future treatment for patients with RCC, concluded Dr. Amin. This is a very small study, but approaches like this are required to move the field forward beyond initial VEGF TKI therapy. Agents with low toxicities, such as immunotherapies, are needed given the prolonged survival rates in RCC in the modern era. It is well known that patients with RCC have developed immune tolerance and increases in suppressor cells that blunt their ability to delay tumor progression. While sunitinib has been shown to reduce this tolerance in certain patients, further strategies to increase this immune response are needed. Further study of this vaccine in controlled trials is warranted.