Mark A. Socinski, MD, Lineberger Comprehensive Cancer Center
University of North Carolina, Chapel Hill, North Carolina

Multitargeted tyrosine kinase inhibitors (TKIs) that inhibit the vascular endothelial growth factor receptor (VEGFR) and other kinases have been evaluated as single agents in advanced refractory non-small cell lung cancer (NSCLC).^4,6,41 Several of these agents have also been studied in combination with chemotherapy.^42-45 Both sunitinib and sorafenib have recently been shown to have single-agent activity in advanced NSCLC. Table 4shows the multiple kinases inhibited by sunitinib and sorafenib. As shown, sunitinib is a more potent inhibitor of both VEGFR and the platelet-derived growth factor receptor (PDGFR) than is sorafenib, but has no activity against the Raf and RET kinases, which are inhibited by sorafenib.

Sunitinib (SU11248)

Sunitinib, an oral multitargeted TKI with antitumor and antiangiogenic activity, inhibits VEGFR, PDGFR, c-Kit, Flt-3, and RET. Sunitinib is approved in the United States for the treatment of advanced renal cell carcinoma and imatinib-refractory gastrointestinal stromal tumors. The efficacy and safety of single-agent sunitinib in previously treated advanced NSCLC were recently evaluated in a multicenter phase II trial summarized in Table 5 ^4,41 and are illustrated as a waterfall plot in Figure 6on the next page. There were no selection criteria with regard to histology, but patients with brain metastasis and baseline hemoptysis were excluded. Patients were treated in two sequential treatment cohorts: 50 mg/day for 4 weeks followed by 2 weeks off (n=63) or 37.5 mg/day continuous dosing (n=47). Both cohorts continued treatment as long as it was clinically beneficial. In the 4 weeks on/2 weeks off cohort (4/2), median age was 60 years (range, 33-86 years), and adenocarcinoma was the most frequent histology (64% vs. squamous cell 22%). Sixty-five percent of the patients were male, and all patients had a performance status (PS) of 0-1. Median number of prior treatment regimens was two (range, 1-4). The primary endpoint was objective response rate (ORR), which was 11.1% (95% confidence interval, 4.6-21.6). In addition, the rate of stable disease (SD) was 44.4%, for an overall clinical benefit rate (ORR + SD) of 55.5%. Progression-free survival (PFS) was 11.3 weeks; overall survival was 23.9 weeks. Most toxicities were grade 1-2, including fatigue/asthenia (43%), pain/myalgia (43%), nausea/vomiting (43%), and stomatitis/mucosal inflammation (21%). Five cases of grade 3-4 neutropenia and thrombocytopenia were reported; no febrile neutropenia was seen. Three hemorrhage-related deaths occurred in the 4/2 arm; two of the deaths were considered to be related to study drug and occurred in patients with squamous cell histology.

Patient characteristics in the continuous dosing cohort were similar (median age, 60 years; age range, 37-81 years; male sex, 57%; adenocarcinoma, 53%; and PS 0/1/2: 49%/49%/2%, respectively).⁴ Again, patients with brain metastasis and gross hemoptysis were excluded. Median number of 4-week cycles administered was three (range, 1-10). Similarly, most toxicities were grade 1-2, including fatigue/asthenia, pain/myalgia, nausea/ vomiting, diarrhea, dyspnea, and stomatitis/mucosal inflammation. Grade 3 or higher adverse events included fatigue/asthenia (15%), hypertension (6%), hypoxia (6%), dyspnea (4%), and hemoptysis (2%). Treatment-related serious adverse events included congestive heart failure (CHF) (1), gastrointestinal bleed (1), hypomagnesemia (1), and hypoxic respiratory failure (1). One patient died due to possible treatment-related CHF. Efficacy data are not yet available.

Response rates for single-agent sunitinib in refractory NSCLC are similar to those for other currently approved agents (e.g., docetaxel, pemetrexed, and erlotinib). Trials that examine sunitinib in combination with standard cytotoxic agents are ongoing and include phase I trials combining sunitinib with carboplatin/paclitaxel as well as cisplatin/gemcitabine. In addition, sunitinib is being evaluated in combination with erlotinib, which is an attractive all- oral regimen targeting the epidermal growth factor pathway (EGFR) in addition to the targets of sunitinib. The latter regimen is being tested in a ran- domized phase II design wherein patients are randomized to erlotinib alone vs. the combination of erlotinib plus sunitinib in the second-line setting. Although sunitinib is active as a single agent, it may be best used in combination with chemotherapy. One exception could be the concept of maintenance sunitinib following first-line chemotherapy, where this well-tolerated oral agent could prolong time to progression. Although the concept of maintenance therapy remains theoretical in advanced NSCLC, an orally administered well-tolerated agent, such as sunitinib, is uniquely suited to address this question.

Sorafenib (BAY 43-9006)

Sorafenib, an oral multitargeted TKI with antitumor and antiangiogenic properties, inhibits Raf, RET and PDGFR, in addition to VEGFR. Sorafenib is approved in the United States for the treatment of advanced renal cell carcinoma. In a multicenter phase II trial examining the efficacy and safety of single-agent sorafenib in advanced NSCLC, patients were treated with 400 mg bid on a continuous schedule and evaluated for response every 8 weeks.⁶ Results are summarized in Table 5. Patients with all histologies were included, as were patients with treated and stable brain metastasis. Fifty-two patients were treated; 51 patients were evaluable for efficacy. Although no partial responses were confirmed, several patients had dramatic cavitary response, documenting activity of this agent in NSCLC (Figure 7). SD response rate was 59%; PFS in patients with SD was 23.7 weeks. Median PFS and median overall survival in evaluable patients were 11.9 and 29.3 weeks, respectively (Figure 8). The most frequently reported drug-related adverse events were diarrhea (40%), hand- foot skin (HFS) reaction (37%), fatigue (27%), and nausea (25%). The most common grade 3-4 drug-related adverse events were HFS (10%) and hypertension (4%). The authors concluded that sorafenib (400 mg bid) was generally well tolerated and showed promising efficacy in patients with refractory advanced NSCLC.

Sorafenib has also been examined in combination with carboplatin/paclitaxel, albeit in a phase I subset analysis.⁴⁵ This analysis was conducted to evaluate the safety and preliminary antitumor activity of oral sorafenib (100, 200, or 400 mg bid) on days 2-18 in combination with carboplatin (area under the curve = 6)/paclitaxel (225 mg/m2) every 21 days in 15 patients with advanced NSCLC. Most drug-related adverse events were grade 1-2 (53%) and none was grade 4. The most common drug- related adverse events were dermatologic (HFS, 20%; rash, 60%) and gas- trointestinal (diarrhea, 20%; anorexia, 13%). Of 14 evaluable patients, 4 patients (29%) had a confirmed PR as best response, 5 patients (50%) had SD, and 3 patients (21%) had PD. Median PFS was 34 weeks. This subset analysis suggests that sorafenib in combination with carboplatin/paclitaxel is tolerable and shows promising antitumor activity in patients with advanced NSCLC. A phase III trial evaluating carboplatin/ paclitaxel with or without concurrent sorafenib in NSCLC is ongoing.

Conclusions

Data from the above-described phase II trials certainly suggest a role for VEGFR kinase inhibitors in the management of advanced refractory NSCLC. Single-agent response rates for these compounds range from 0% to11.1%, and SD rates seem to consistently range between 40% and 59%.4,6,41-44In general, these agents are well tolerated, but toxicity related to CNS and pulmonary hemorrhage needs to be further defined, particularly in patients with squamous histology and/or CNS metasta- sis. The optimal use of antiangiogenic agents is more likely in combination with standard cytotoxic or other targeted agents; however, the most effective combination with the least toxicity is yet to be determined. The challenge ahead is to identify the most useful applications for these agents and to unravel which targets are most important for individual patients.

Discussion

Dr. Lynch: Are sunitinib and sorafenib really different classes of drugs that have been lumped together principally because they've both been developed for renal cell carcinoma and the approvals have been close in time? Or, should we be thinking of them as different drugs?

Dr. Kim: The two drugs have been lumped together because of the coincidental nature of their approvals in renal cancer. However, both drugs have other targets in addition to VEGF, so you would expect their side effect profiles and mechanisms to be a little different.

Dr. Lynch: Do you use sunitinib and sorafenib as single agents yet in lung cancer?

Dr. Socinski: I have used sunitinib sparingly in a few patients who have exhausted standard options, yet still have a good performance status. I was more impressed by the sunitinib waterfall plot, as it shows a number of minor responses where patients derived clinical benefit, which is notable in this particular setting.

Dr. Lynch: What is the value of waterfall plots to a community oncologist?

Dr. Socinski: The visuals are valuable because I think that all of us struggle with the category of what we call "stable disease" and the nuances of the Response Evaluation Criteria in Solid Tumors Group (RECIST) classification criteria for responding disease. It's very difficult in refractory populations to know whether an agent will have utility. The sunitinib waterfall plot gives you some idea of antitumor activity.