We've received several questions about agents that might be helpful for patients who have already responded to inhibitors of the epidermal growth factor receptor (EGFR) like Tarceva (erlotinib) and Iressa (gefitinib) and then demonstrate progression. These latter agents are reversible inhibitors of of the tyrosine kinase domain (signalling portion inside the cell) of the EGFR molecule, meaning that they attach to and periodically detach from the receptor. Other inhibitors, like the novel Pfizer agent PF299804, bind to EGFR irreversibly, never coming off of the receptor, and requiring the cell to make new EGFR molecules without an inhibitor on them. Such agents can kill many kinds of cancer cells in a lab-based model, and appear that they may do so more effectively than currently agents like Tarceva and Iressa, but how well they work in real patients has remained an open question. Another unresolved issue is whether PF299804, an inhibitor of not only EGFR but of other members of the human epidermal growth factor receptor (HER) family, of which EGFR (also known as HER1) is just one type, are more effective in patients than agents that inhibit EGFR alone. Such agents that block multiple members of the HER family are sometimes referred to as "pan-HER" inhibitors (as in "across the HER family"), but they're still in clinical studies to determine whether such agents provide incremental benefit beyond what we see with the EGFR-specific agents we already use. Though results with the orally available irreversible pan-HER inhibitor PF299804 weren't a lead story at ASCO 2010, I think several of these trials were quite encouraging, both for patients with an EGFR mutation who might seek something after they become resistant to an EGFR inhibitor that previously was very beneficial, and also for people who don't have an EGFR mutation and hope to do better than they might expect to do with an agent like Tarceva or Iressa.

A pair of studies were done on patients with advanced NSCLC who had previously been pretty treated with both chemotherapy and an EGFR inhibitor. First, one done in Asia by Park and colleagues enrolled 12 patients in an initial phase I portion, followed by an additional 42 patients in the subsequent phase II study after a dose of 45 mg by mouth daily had been identified as the appropriate target dose. The analysis focused on the patients in the phase II portion at the 45 mg daily dose, of whom 40 had response data available. Overall, these results were quite encouraging, with 48% of patients still without progression 4 months after starting treatment, 15% showing a partial response (PR), and another 52.5% demonstrating stable disease (SD) as their best response. The duration of treatment is shown in the figure below, with the length of the horizontal line extending to the right proportional to the duration of time on the drug and without progression: (click on image to enlarge) The waterfall plot below shows the distribution of tumor shrinkage (proportional to the length of the lines going down from the horizontal line, clustered to the right), progression (upward lines clustered to the left), and stable disease (short lines not going much down or up): Of course, the other important factor is tolerability, which was pretty good. The leading side effects were diarrhea, rash, and mouth sores, all typically in the mild to moderate range. A very similar trial of American patients was reported by Campbell and colleagues. This study enrolled 66 patients with advanced NSCLC, previously treated with both chemotherapy and erlotinib, and over 80% had tumor tissue available for molecular marker studies. They divided their population into the 50 with an adenocarcinoma, of whom more than half were never-smokers and the majority had an EGFR mutation, and the other 16 with a non-adenocarcinoma, who were much less likely to be never-smokers or have an EGFR mutation. They saw a wide range in duration of treatment, with 13 patients (23%) demonstrating prolonged clinical benefit (either a complete response (CR), PR, or SD lasting at least 24 weeks). Not all of these patients had an EGFR mutation (7 did, 4 were EGFR wild type (no mutation), and 2 were unknown). Overall, though, progression-free survival was longer in patients who had an EGFR mutation: The side effect profile was remarkably similar to the Asian experience, though the investigators noted that the side effects seemed to gradually decrease with ongoing treatment. Overall, then, these studies both demonstrate that an encouraging fraction of pretty extensively treated patients with advanced NSCLC show tumor shrinkage or at least prolonged stable disease with this oral agent, and that the results, while perhaps most encouraging in patients with an EGFR mutation or the demographic features where it is prevalent, convincingly reveal activity in patients who don't have an EGFR mutation. How does PF-299 stack up against erlotinib in patients who haven't received an EGFR inhibitor previously? I'll cover that next.