Sequential therapy in renal cell carcinoma†
This educational proceedings publication is based on a symposium held on June 27-28, 2008, in Cambridge, Massachusetts.
Because of the recent approval of several drugs for the treatment of renal cell carcinoma (including sorafenib, sunitinib, temsirolimus, and, in Europe, bevacizumab plus interferon), the use of sequential therapy has become routine practice. There is now evidence that administering these targeted agents sequentially provides clinical benefit by inducing tumor shrinkage and prolonged progression-free survival (PFS) in a large number of patients. However, data regarding overall survival (OS) are still pending. By adding these drugs in an adequate order, one can expect an increase in overall PFS of up to 27 months and a subsequent improvement in the OS of patients with renal cell carcinoma. It has been recently reported that the OS of patients treated with sunitinib in the first-line setting was 26 months. Expecting a survival of 40 months does appear possible based on currently available data, although this assumption will have to be proven in the future. Cancer 2009;115(10 suppl):2321-6. © 2009 American Cancer Society.
The approval of several drugs for the treatment of advanced renal cell carcinoma (RCC) has provided many therapeutic options for this disease. After the approval of interleukin-2 (IL-2) in the 1980s (and of interferon in many countries), subsequent approval (based on data from large, randomized, phase 3 studies) has been given to sorafenib,1 sunitinib,2 and temsirolimus3 in the US and Europe. In addition, bevacizumab in combination with interferon4 was approved in December, 2007 in Europe, and everolimus (RAD-001) has recently demonstrated activity in patients in whom previous treatment with tyrosine kinase inhibitors (TKIs) has failed, raising the expectation that this drug might be approved in 2009.5
Obviously, the availability of these drugs encourages oncologists but also patients to try each drug sequentially, at least in those countries in which there is access to these agents. This review will provide available data regarding sequential therapy and will attempt to draw hypotheses for the future therapy of RCC. Although there is evidence that sequential therapy can provide some benefit in terms of disease control and tumor shrinkage, to our knowledge, a survival benefit has not yet been proven.
Efficacy of Targeted Agents After Cytokine Therapy
Two randomized studies have demonstrated the efficacy of targeted agents versus placebo in patients with cytokine–refractory disease. Bevacizumab was first shown to be active in patients in whom high-dose IL-2 treatment had failed. In this randomized phase 2 trial,6 116 patients were randomized to 1 of 3 treatment arms: placebo (n = 40 patients), 3 mg/kg of bevacizumab (n = 37 patients), or 10 mg/kg of bevacizumab (n = 39 patients). Data from this trial demonstrated that the median time to disease progression was significantly longer for the patients treated with 10 mg/kg of bevacizumab than placebo (4.8 months vs 2.5 months; hazards ratio [HR], 2.55 [P <.001]). The median time to disease progression for the patients receiving 3 mg/kg of bevacizumab was 3.0 months and was not significantly greater than that of patients on the placebo arm (HR, 1.26; P = .053). Four patients (10%) treated on the 10-mg/kg treatment arm achieved partial responses, which were of variable duration (6 months, 9 months, 15 months, and >39 months, respectively). No patients on the 3-mg/kg treatment arm achieved a partial response. Overall survival (OS) was not found to be significantly different among the 3 treatment arms (P >.20 for all comparisons), which could be due to the crossover of the placebo patients.
Sorafenib also demonstrated activity in a large, randomized, phase 3 trial.1 In this 903-patient study, patients were randomized to receive either sorafenib (at a dose of 400 mg administered twice daily) or placebo. Tumor shrinkage was observed in the majority of patients, and progression-free survival (PFS) significantly improved from 3 months to 6 months (P <.0001). Final OS was reported in 2007.7 Although the OS was found to be similar between the 2 treatment arms (17.8 months vs 15.2 months; P = .15), there was a suggestion of a preplanned analysis demonstrating improved survival with sorafenib after the censoring of placebo patients who crossed over to the sorafenib arm (17.8 months vs 14.3 months; P = .03).
Several other targeted agents also demonstrated activity in patients with cytokine–refractory disease. Sunitinib has the strongest evidence of activity outside of randomized phase 3 trials; in 2 consecutive phase 2 studies, for a total of 160 RCC patients with metastatic RCC who failed first-line, cytokine-based therapy, the response rate was approximately 40%, with a median PFS of 8 months and an OS of approximately 16 months.8, 9 Axitinib, another TKI, also demonstrated impressive activity in a 52-patient, phase 2 study in patients with cytokine–refractory RCC, with a 44% objective response rate and a median time to disease progression of 15.7 months.10 Finally, in an initial randomized phase 2 study designed to determine the best dose for further phase 3 studies, temsirolimus was found to result in a 5.8-month PFS, mainly in patients in whom first-line therapy with cytokines had failed.11
Thus, all currently tested targeted agents have activity after cytokines. Although the magnitude of PFS (Table 1) appears to be different between agents (ranging from 4.8 months–15.7 months), these numbers should be viewed with caution because the trial designs and the selected populations might bias them considerably.1, 6, 8-11
Table 1. Efficacy of Targeted Agents After Cytokine Therapy
|Bevacizumab6||Randomized phase 2||110||4.8|
|Sorafenib1||Randomized phase 3||904||6|
|Sunitinib8, 9||Phase 2||168||8|
|Temsirolimus11||Randomized phase 2||111||5.8|
Conversely, the efficacy of cytokines after targeted agents remains unknown, although toxicity might increase in this setting as reported for IL-2 after targeted agents.12
Efficacy of TKIs After Vascular Endothelial Growth Factor–directed Therapy
To our knowledge, no randomized study has assessed the efficacy of TKIs after either bevacizumab or TKIs. However, several prospective studies and a few retrospective analyses of patients treated sequentially with TKIs support the finding that there is no absolute cross-resistance among those drugs that target vascular endothelial growth factor (VEGF).
Sunitinib has demonstrated activity in patients who had been previously treated with bevacizumab and subsequently experienced disease progression.13 That study evaluated the utility of sunitinib, at a classic dose of 50 mg/day for 4 weeks on and 2 weeks off, in 61 patients who received at least 4 doses of bevacizumab and demonstrated Response Evaluation Criteria in Solid Tumors (RECIST)–defined progression of disease during or within 3 months of bevacizumab-based treatment. The best response reported was 23% of patients achieving a partial response and 74% of patients overall demonstrating some degree of tumor shrinkage.13 The median PFS was 29.7 weeks. Three patients achieved a partial response with both agents, whereas other patients achieved a partial response only with bevacizumab (n = 2 patients) or sunitinib (n = 11 patients), without an objective response to the other agent.13
Axitinib has demonstrated promising efficacy in patients with disease that is refractory to sorafenib.14 In a phase 2 study, 62 patients received axitinib at a daily dose of 5 mg, with a partial response rate of 21% and a PFS of 7.4 months reported.14 It is interesting to note that, in this study, the patients could have received other therapies, including sunitinib and bevacizumab, before sorafenib was administered. It was discovered that many patients received sorafenib in the third-line setting or later, outlining the promising efficacy of axitinib in RCC.14
Finally, several reports of retrospective analyses of patients sequentially treated with sorafenib and sunitinib or vice versa have been published to date.15-17 What to our knowledge is the largest study reported to date, consisting of 90 consecutive patients,15 was reported recently. In this study, 68 patients received sorafenib followed by sunitinib, whereas 22 patients received the reverse sequence. Benefit was observed in both groups with the second TKI, with tumor shrinkage and partial responses reported in some patients. The PFS times were 25 weeks and 17 weeks, respectively for sunitinib and sorafenib given as the second TKI, with an overall median duration of combined therapy of 61 weeks and 49 weeks, respectively, for the sorafenib-sunitinib and sunitinib-sorafenib sequences.
Efficacy of mTOR Inhibitors After TKIs
Until recently, no data have been available regarding the use of mTOR inhibitors after TKI therapy. Recently, Motzer et al5 reported the results of a randomized study of everolimus versus placebo in patients with TKI–refractory disease. Patients with RCC with a clear cell component that progressed ≤6 months after VEGF receptor (VEGFR)–TKI therapy (sorafenib, sunitinib, or both) were randomized 2:1 to treatment with everolimus (10 mg/day orally) or placebo, both of which were administered with best supportive care. Patients were stratified by Memorial Sloan-Kettering Cancer Center risk criteria and prior VEGFR-TKI therapy (1 vs 2 regimens). The primary endpoint was PFS, which was documented using RECIST and assessed via blinded, independent review. At the time of disease progression, treatment was unblinded, and patients receiving placebo were offered open-label everolimus. A total of 272 patients were randomized to receive everolimus and 138 were randomized to receive placebo. The PFS was 4 months versus 1.9 months according to the independent review, leading to a significant improvement (HR, 0.31; P <.001). This benefit was observed in all TKI groups: sunitinib alone (46%), sorafenib alone (28%), or both TKIs (26%). In addition, tumor shrinkage was observed in 60% of patients versus 10% with everolimus therapy compared with placebo.
The efficacy of everolimus in combination with bevacizumab after TKI has also been recently reported.18 In a phase 2 study, 59 patients received this combination therapy, 29 of whom had been previously treated with sunitinib or sorafenib. PFS in this patient population was 10 months, suggesting the synergistic activity of both agents in patients with TKI–refractory disease.
Speculations and Suggestions for the Sequential Treatment of RCC
To our knowledge, the data reported herein are currently the only available data to support sequential therapy in RCC. Data are not available for either bevacizumab or temsirolimus after TKIs, nor for the administration of mTOR inhibitors after another mTOR inhibitor. Thus, deciding which is the best sequence to use in patients with RCC remains hypothetical.
An attempt to speculate based on the data was made in Table 2. In this table, the calculation was made by adding the PFS obtained from reported studies to provide an expected PFS in patients able to receive all the active drugs sequentially (which is obviously biased). Although not scientifically valid, the findings in Table 2 suggest that one could obtain an “overall PFS” ranging from approximately 20 months to 27 months by sequential therapy.1, 2, 4, 10, 12-15 Based on recent reports that sunitinib (with a median PFS of 11 months2) induced an OS of 26 months19 whereas interferon (with a median PFS of 5 months) provided a median OS of 15 months, one can expect that, with a total PFS of approximately 27 months using the best sequential therapy, the survival of patients with RCC might increase up to ≥40 months in the near future. We recognize that this is pure speculation. However, this speculation raises interesting hypotheses for future clinical trials.
Table 2. Progression-free Survival of Patients Receiving Sequential Therapy*
|Sunitinib2||Sorafenib||Everolimus15|| ||Total PFS, Months|
|Bevacizumab with interferon4||Sorafenib1||Sunitinib14||Everolimus15|| |
|Bevacizumab with interferon4||Sunitinib12||Sorafenib14||Everolimus15|| |
|Interferon2||Axitinib10||Everolimus15|| || |
|Bevacizumab with interferon4||Sorafenib1||Axitinib13||Everolimus15|| |
In conclusion, there is evidence that sequential therapy in RCC is beneficial to patients and ultimately should provide a survival advantage. This hypothesis should now be demonstrated by prospective, well–designed trials. Therefore, future trials should focus on the efficacy of 1 sequence compared with another. Some interesting proposals are currently under evaluation, such as the University of Texas M. D. Anderson Cancer Center “START” proposal. In this proposal, RCC patients are randomized at the time of first–line therapy between bevacizumab, sunitinib, or temsirolimus; after disease progression, the same patients are subsequently randomized between the 2 remaining drugs (patients first treated with bevacizumab are randomized between sunitinib and temsirolimus, etc). Another proposal comparing everolimus followed by sunitinib with sunitinib followed by everolimus will raise another sequential question.
Finally, it is likely that a better understanding of the mechanisms underlying efficacy but also resistance to these targeted agents will help in the future. As an example, the evidence that hypoxia–inducible factor expression into the tumor is strongly associated with response to sunitinib20 suggests that the biology of the tumor will be important for selecting the best agent with which to initiate treatment.
The questions and discussion below follow from the oral presentation given at the Third Cambridge Conference on Innovations and Challenges in Renal Cancer and do not correspond directly to the written article, which is a more general review.
Dr. Michael Atkins: Did patients with prior sunitinib therapy who demonstrated activity with sorafenib not tolerate sunitinib or not have progressive disease?
Dr. Bernard Escudier: The best response to sunitinib was progressive disease. I would not consider using sorafenib in patients who had progressive disease while taking sunitinib.
Dr. Brian Rini: In our trial of sorafenib after sunitinib, we did not see an association between response to the first therapy and response to the second therapy. Obviously, if someone blows through their first line of therapy, they may not be in good shape to do well with other agents, but I do not think that lack of response on sunitinib precluded lack of tumor shrinkage with sorafenib.
Dr. Escudier: When we don't have at least stable disease as best response, efficacy of sorafenib has not been seen in our experience.
Dr. Robert Figlin: I am not yet clear on how to design trials where you have the potential for sequential therapy leading to an overall survival of almost 3.5 years where responses sometimes can be modest. Progression-free survival requires randomized trials but the industry does not want to compare their drugs to each other. I worry that in 5 years we will have more information but no clear path to the top.
Dr. Escudier: I want to have a combination trial with complete remission rate as a primary endpoint.
Dr. Figlin: Are there laboratory-based questions that will help us determine what we need to be thinking about in the clinic in terms of generating resistance in the clinic and then adding other agents?
Dr. William Kaelin: We are continuing to try to identify additional targets that make sense because either they are on the HIF pathway or they are synthetic lethal to VHL. However, I do not know of a reason in the laboratory why the mTOR inhibitors should not have an additive effect with the VEGF inhibitors. Despite this, I am sensing some pessimism from the community about whether mTOR inhibitors and VEGF inhibitors are going to be better than monotherapy.
Dr. Atkins: Let us say we had a combination in the BeSt trial that was better than bevacizumab. One question would be whether that combination would be better than sequential therapy.
Dr. Kaelin: Maybe we need a new paradigm, such as sequential rather than combinations. On the other hand, combinations are important for those forms of cancer we can cure with chemotherapy. Maybe for epithelial cancers like kidney cancer, we will need 3 or more active drugs.
Dr. Figlin: If you had 3 or 4 drugs, all of which produced partial remissions and none of which produced complete remissions, you still couldn't cure anyone. The reason you cure people with Hodgkin disease and lymphoma is because you produce complete remissions.
Conflict of Interest Disclosures
The program was made possible by educational grants provided by Genentech, Novartis Pharmaceuticals, Pfizer, Inc, and Wyeth Pharmaceuticals. Program management and CME sponsorship were provided by InforMEDical Communications, Inc, Carlisle, Massachusetts.
Bernard Escudier has received honoraria from Bayer, Hoffman LaRoche, Genentech, GlaxoSmithKline, Novartis, and Wyeth.