Abstract
Background/Aim: the ideal sequence of targeted agents for advanced kidney cancer is still unknown. In the present study we assessed the clinical benefit of two different sequential approaches, namely sorafenib, an inhibitor of mammalian target of rapamycin (mTORi) and sunitinib, or sunitinib (an mTORi) and sorafenib. Patients and Methods: we retrospectively reviewed the outcome of 40 advanced kidney cancer patients treated with one of the two above sequences. Results: a total of 26 patients were treated with the sequence sorafenib-mTORi-sunitinib and 14 with the sequence sunitinib-mTORi-sorafenib. The actuarial overall median progression-free survival (PFS) in the sorafenib-mTORi-sunitinib group and in the sunitinib-mTORi-sorafenib group were 21.9 and 22.8 months, respectively (log-rank test: p=0.928). In the sorafenib-mTORi-sunitinib group, patients in first-, second- and third-line therapy experienced PFS of 11.7, 5.1 and 9.1 months, respectively, while in the sunitinib-mTORi-sorafenib group PFS was 14.4, 4.3, and 3.9 months, respectively. Conclusion: Our results suggest there is no significant difference between the two sequence modalities.
The recent availability of six drugs aimed at specific molecular targets for the treatment of metastatic renal cell carcinoma (RCC) (1), and the high percentage of disease control rate obtained with each of these drugs in clinical practice has meant that a sequential approach to treatment has often been adopted (2).
Data derived until recently from a limited number of prospective studies and from a larger number of retrospective studies (although these latter were compromised by inevitable selection bias) would seem to support the sequential use of tyrosine kinase inhibitors (TKIs), whose cross-resistance appears to be limited (3). Such a hypothesis was recently confirmed by the results from the axitinib registrative trial (4).
On the other hand, the results of the RECORD-1 study showed that the use of a drug with a putatively different mechanism of action, the mammalian target of rapamycin inhibitor (mTORi) everolimus, is able to significantly prolong progression-free survival (PFS) compared to placebo in patients pre-treated with sunitinib, with sorafenib, or with both TKIs. The extent of the reduction in risk of progression observed with the use of everolimus after one or both TKIs also appears to be quite similar (5, 6).
These observations lead us to the important problem of which sequence of these three drugs can achieve the most significant benefit in terms of overall PFS: TKI-TKI-mTORi or TKI-mTORi-TKI (3).
The results of the axitinib registrative study have to a certain extent provided balanced evidence for the two possible options, although the problem of confirming the efficacy of the TKIs after use of the mTOR inhibitors remains. In contrast, the efficacy of everolimus after TKI is supported by data from the RECORD-1 study '95 mentioned earlier.
Only one small retrospective study has explored the activity of a TKI (namely, sorafenib) administered after a first TKI followed by an mTORi. This study suggested the feasibility and above all the efficacy in terms of prolonged PFS for this sequential approach (7).
The aims of the present study were, therefore, to confirm the efficacy of the use of a TKI after an mTORi and to compare the benefit in terms of overall PFS of the two sequential approaches: sunitinib-mTORi-sorafenib and sorafenib-mTORi-sunitinib. The retrospective nature of this analysis makes the transfer of results difficult and this consists a limitation. However, in spite of this, this study aims to generate hypotheses on this important clinical problem.
Patients and Methods
Patients. We performed a retrospective analysis of 40 patients with advanced RCC treated between September 2005 and October 2010 at six European centers.
This retrospective study was approved by the Ethical Committee of the I.R.C.C.S. San Matteo University Hospital Foundation.
All patients had received first-line treatment with either sunitinib (50 mg daily, four weeks on and two weeks off) or sorafenib (400 mg twice daily, continuous dosing), followed by a second-line treatment with an mTORi (either 10 mg everolimus daily continuous dosing or 25 mg temsirolimus intravenous weekly) and, upon further progression, with the other multikinase inhibitor (sorafenib or sunitinib), as third line therapy.
Response assessment. The status of disease progression during the three treatment periods was determined by radiological assessment using the Response Evaluation Criteria In Solid Tumors (RECIST) (8) approximately every 12 weeks. Patients who were treated with any other agent during the gap between the three drugs treatments were excluded from this analysis.
Overall PFS evaluation. The primary end-point of this analysis was to determine overall PFS, defined as the sum of the PFS achieved under each of the three drugs, excluding any time which may have elapsed between each treatment period.
Data analyzed in this study were obtained from the medical records of each individual patient; baseline characteristics, date of start of treatment, dates of progression, and time between these treatments were all recorded.
PFS for the first line therapy was calculated as the time from the start of the first TKI to the time of disease progression on the same drug. PFS for the second line was calculated as the time from the start of an mTORi to the time of disease progression on the same drug. Finally, PFS for the third line was calculated as the time from the start of the second TKI to the time of disease progression or death.
Patients who remained on treatment on the second TKI without disease progression at the end of the study period were censored.
Statistical analysis. Quantitative variables were not normally distributed (Shapiro-Wilk's test) and hence the results are expressed as median values and interquartile range (IQR; 25th-75th percentile). Comparisons between the two treatment groups were performed by the Mann-Whitney test. χ2 statistics or Fisher's exact test, as appropriate, were applied to compare qualitative variables. PFS was expressed as median and IQR, and was analyzed by the Kaplan Meier method, while Cox's regression models were used to analyze associations between PFS and baseline characteristics and treatment groups. Results were expressed as hazard ratios (HR) with their 95% confidence intervals (95% CI). A value of p<0.05 was considered statistically significant. All tests were two-sided. Data analysis was performed with the STATA statistical package version 10 (Stata Corporation, College Station, TX, USA).
Results
Patients. A total of 40 patients were included in this retrospective analysis. Twenty-six patients were treated with the sequence sorafenib-mTORi-sunitinib and 14 with the sequence sunitinib-mTORi-sorafenib.
As far as the mTORi-based second-line treatment is concerned, in the sorafenib-mTORi-sunitinib sequence, all patients received everolimus, while in the sunitinib-mTORi-sorafenib sequence, 5 out of 14 patients (i.e., 35.7%) received temsirolimus and the remaining were treated with everolimus.
Patients' baseline characteristics are summarized in Table I. These were similar for both patient groups in terms of age (p=0.487), performance status (p=0.973), Motzer's score (p=0.431), Fuhrman's grade (p=0.429), and presence of liver metastases (p=0.507). In the sunitinib-mTORi-sorafenib group, a higher incidence of non-clear cell metastatic RCC was observed (5 out of 14 patients vs. 0 out of 26 in the sorafenib-mTORi-sunitinib group; p=0.001).
PFS after first-line treatment. On completion of first-line treatment, no statistically significant difference in terms of PFS was observed between patients who received sorafenib or those who recevied sunitinib (median PFS=11.67 (6.08-19.57) and 14.45 (7.14-20.26) months, respectively; p=0.8768). Relevant PFS curves are shown in Figure 1.
PFS after second-line treatment. As far as second-line treatment is concerned, consisting of an mTORi for both groups, no statistically significant difference was observed (p=0.9492) in terms of PFS independently of the TKI used as first-line treatment: Sorafenib: median PFS=5.06 (3.90-6.74) months and sunitinib: median PFS=4.26 (2.08-19.07) months. Relevant PFS curves are shown in Figure 2.
Baseline demographic and clinical characteristics.
PFS after third-line treatment. Those patients who had received sorafenib as third-line treatment had a longer PFS compared with that observed for patients treated with sunitinib [median PFS=9.12 (3.50-20.03) and 3.90 (3.00-13.42) months, respectively; p=0.2379]. Relevant PFS curves are shown in Figure 3.
Overall PFS after complete sequential treatment. Among patients treated with sorafenib-mTORi-sunitinib or sunitinib-mTORi-sorafenib, at the time of data lock (30-10-2010), 7 out of 26 (i.e., 26.9%) and 4 of 14 (i.e., 28.6%) patients, respectively, were still under active treatment.
There was no statistical difference in the actuarial overall median PFS (not including inter-treatment periods) between the sorafenib-mTORi-sunitinib group and the sunitinib-mTORi-sorafenib group [21.9 (17.0-39.6) months vs. 22.8 (13.9-29.4) months; p=0.9282]. Relevant PFS curves are shown in Figure 4.
Univariate analysis. As summarized in Table II, in the univariate analysis, age, sex, non-clear cell histology, ECOG performance status, Motzer's score, Fuhrman's grade and the presence of hepatic metastases were not associated with overall PFS.
Discussion
In spite of the unquestionable improvements achieved in the treatment of RCC with the introduction of molecular-targeted drugs, prognosis of patients with advanced disease can and must be further improved. In order to achieve this, two different approaches seem to be possible: the combination of different drugs or their sequential use (9). However, various reasons have been given against the use of a combination of two or more molecular-targeted drugs. Firstly, doubts remain as to whether a significant improvement can be achieved. Secondly, the problem of toxicity which appears to be more than cumulative remains. Finally, economic constraints make the use of these associations unsustainable (9, 10).
Influence of baseline characteristics on overall survival (univariate analysis).
Available data clearly suggest that sequential therapy is feasible and efficacious (2, 3, 9, 11). In spite of this, it is still not known which sequence is best able to provide the greatest benefit in terms of overall survival or, at least, in PFS.
The optimal treatment for patients whose disease progresses after first-line treatment with a vascular endothelial growth factor (VEGF) pathway inhibitor remains an unresolved question. Available data so far seem to suggest the possible use of either a second drug of the same class or of a different drug, such as an mTORi, as shown by randomized phase III studies of axitinib (4) and everolimus (5, 6), and by numerous, mainly retrospective studies of the sequential use of sunitinib and sorafenib (3). An attempt has recently been made to rationalize and personalize the choice of therapy in second- and third-line treatments taking into consideration the drug activity and tolerance observed in first-line treatment (3).
Data from the RECORD-1 study clearly show that the efficacy of the mTORi everolimus does not depend on whether it was used after one rather than two TKIs (5, 6). In contrast, the action of a VEGF pathway inhibitor, when used after the failure of a second-line treatment with an mTORi, is currently only supported by a small retrospective study (7).
Our results are also derived from a retrospective study and are, therefore, compromised by an unavoidable selection bias. However, they do point in the same direction as data from Di Lorenzo et al. (7) which suggest that patients may be sensitive again to a TKI after a second-line treatment with an mTORi. Interestingly, both the prospective AXIS data (4) and the retrospective reports on the action of a rechallenge with both sunitinib and sorafenib (12, 13), as well as a recent report by Iacovelli et al. (14), do suggest that a continuous inhibition of the VEGFR pathway could be of benefit for patients with advanced kidney cancer. The activity of everolimus in patients failing a first- or second-line treatment with an anti-VEGF pathway TKI does not necessarily contradict this hypothesis. Indeed, due to the particular molecular pathogenesis of kidney cancer, which is so dependent on angiogenesis due to the frequent mutation of the Von Hippel-Lindau (VHL) gene ever in sporadic cases (15), it is also possible that currently available mTORi [which are indeed just mTOR complex 1 (mTORC1) inhibitors] could also be active due to their indirect inhibition of angiogenesis (16). A series of ongoing studies will address this key issue. In the meantime, we should probably try to simplify the whole picture (17), optimizing as much as possible the overall treatment management of our patients. Indeed, the issue today is to use currently available agents, well, and not which of these drugs is, per se, the best.
PFS Kaplan–Meier estimates for patients treated with sorafenib (So) or sunitinib (Su) as initial therapy within the two possible sequences (So-mTORi-Su or Su-mTORi-So).
PFS Kaplan–Meier estimates of patients treated with mTORi as second-line therapy within the two possible sequences (So-mTORi-Su or Su-mTORi-So).
PFS Kaplan–Meier estimates for patients treated with sorafenib (So) or sunitinib (Su) as third–line treatment within the two possible sequences (So-mTORi-Su or Su-mTORi-So).
Kaplan–Meier estimates of overall PFS after complete sequential treatment (either So-mTORi-Su or Su-mTORi-So).
Our results suggest there is no significant difference between the two sequential modalities described here, supporting similar findings (18). Treatment should thus be tailored, taking into account patient-, disease- and treatment-related characteristics (19).
- Received August 1, 2013.
- Revision received October 8, 2013.
- Accepted October 10, 2013.
- Copyright© 2013 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
