For many years, chemotherapy for advanced or metastatic NSCLC had been limited to the use of “doublet” (two-drug) therapy using different combination regimens that were overall found to have very similar outcomes, but with different toxicity (side effect) profiles. Attempts to add a third chemotherapy agent for a triplet regimen, and numerous attempts to add different targeted-therapy agents, had dismal success. Not only did most of the combinations fail to improve on the survival outcomes, they increased the number of side effects compared with doublet chemotherapy alone.

Avastin was the first targeted therapy agent to improve upon these outcomes. In a landmark phase III trial conducted by the Eastern Cooperative Oncology Group (ECOG 4599), 878 patients with advanced NSCLC received either standard-of-care carboplatin with paclitaxel, or the same doublet chemotherapy with Avastin. Due to risks of fatal bleeding in patients with squamous cell NSCLC seen in an earlier study, this trial was restricted to patients with non-squamous NSCLC. Patients received 6 cycles of chemotherapy with or without Avastin, and those on the Avastin arm then received maintenance Avastin until disease progression or intolerable toxicity.

Median survival with the doublet regimen was consistent with survival outcomes from previous studies (10.3 months), while survival for patients on the Avastin arm was significantly superior by 2 months (12.3 months). Progression-free survival (PFS, the time it takes for a cancer to start growing on chemotherapy) was also improved with the triplet, with PFS of 6.2 months compared with 4.5 months with the doublet regimen alone.

Addition of Avastin to chemotherapy slightly increased the immune system suppression seen with chemotherapy alone, and added side effects consistently seen with Avastin therapy, including hypertension, proteinuria (the kidneys spilling protein into the urine), bleeding, rash, headache, and low blood sodium levels.

This triplet combination quickly became a standard of care regimen for this select patient population (patients with good performance status, non-squamous cell lung cancer, no brain metastases, etc). Although the regimen is fairly well-tolerated, the regimen does have side effects, including cumulative neuropathy that can significantly affect quality of life for patients and prevents prolonged administration of paclitaxel.

Alimta is a chemotherapy drug with activity against non-squamous NSCLC. In comparison to many other chemotherapy agents, it is fairly well tolerated. When given with appropriate vitamin supplementation (B12 and folic acid), the incidence of severe immunosuppression is very low. It causes less neuropathy, nausea/vomiting and hair loss than many other NSCLC chemotherapy drugs.

Dr. Jyoti Patel and colleagues conducted a phase II trial evaluating the safety and efficacy of the combination of carboplatin, Alimta, and Avastin for patients with advanced non-squamous NSCLC. Similar to the structure of the ECOG 4599 study, patients received 6 cycles of the triplet combination of carbo/Alimta/Avastin, but then received maintenance therapy with both Alimta and Avastin until progression of disease or toxicity (compared with Avastin alone in ECOG 4599).

This was a nonrandomized trial involving 51 patients and was NOT designed to test for superiority of survival. The combination was fairly well tolerated, with the interesting findings that no patient severe bleeding or hypertension as had been seen with previous Avastin studies. Four patients however developed aggravation of previously-known diverticulitis, which caused the investigators to modify the trial to exclude patients with a history of diverticulitis.

PFS in this trial was 7.8 months, and median overall survival was 14.1 months. While there is no comparison arm here, the results are certain encouraging compared with what we expect to see for a similar population. It is important to mention, however, that in phase II trials, overall survival tends to be higher than those of a phase III study. In the initial phase II trial evaluating carbo/paclitaxel/Avastin, the triplet combination demonstrated an overall survival of 17.7 months, far better than the still very good 12 month survival seen in the phase III experience.

Given the favorable toxicity profile and promising survival results, a randomized phase III trial comparing the ECOG 4599 regimen to the “Patel regimen” has been initiated. This trial is a pure head-to-head study of the two regimens to see which may be better in terms of overall survival. PFS, toxicity, and quality of life are also being compared. This is a large trial, with a planned accrual of 900 patients with nonsquamous NSCLC, but it allows patients with treated and stable brain metastases, reflecting the growing evidence supporting the safety of Avastin in this population and the increasing comfort with using Avastin in these patients. As in the prior trials, patients with diverticulitis are excluded, as are patients receiving blood thinners. There are 131 sites in the United States currently participating, and the results will be eagerly awaited.

In the meantime, oncologists are stuck in a quandary about how to treat patients with first-line therapy, and what options to consider in the maintenance setting. For patients who would fit the ECOG 4599 profile, one standard of care is the triplet combination as described above. Yet this combination can be difficult for patients, especially in terms of the neuropathy. We have good data that Alimta is a very effective chemotherapy drug against nonsquamous NSCLC (particularly adenocarcinoma), and has fewer cumulative side effects. Particularly when we have a phase III trial indicating improved outcomes when chemotherapy is chosen based upon histology (with patients with adenocarcinoma having improved survival with cisplatin/Alimta compared with cisplatin/Gemzar), and phase II data showing that the triplet of carboplatin/Alimta/Avastin seems to have an acceptable safety profile, it is difficult to resist offering patients the “Patel regimen” off-study, if clinical trial participation is not available.

Is this the correct choice? We don’t know at this time, and likely won’t for a few years until the phase III study is completed. As an evidence purist, I would stick with the ECOG 4599 regimen for those patients fitting that profile as my first choice. That being said, sometimes evidence purists also have to be realists, and there are patients for whom a triplet would be feasible, but the neuropathy concerns may be significant (particularly when a patient already has a preexisting neuropathy). I would then consider either cisplatin/Alimta based on Dr. Scagliotti’s trial above, or the Patel regimen as triplet therapy. In either case, if a patient has stable disease and is tolerating treatment well, I consider continuing Alimta with or without Avastin in the maintenance setting.

The choices for treating first-line advanced NSCLC are increasing rapidly, and the choices in the maintenance setting (Avastin? Avastin with Tarceva? Alimta? Alimta with Avastin? Taxotere?) are unprecedented. This makes the decisions of how to design the next trials extremely difficult.

There will be no easy answer. The individual clinical situation needs to be carefully evaluated for each patient, as does quality of life throughout all treatments. The future changes in the US healthcare system will likely have a heavy influence on these issues in the very near future, as well, since both Alimta and Avastin are very costly treatments. For now, I am encouraged that our horizons seem to be broadening. Hopefully they will bring daylight soon.