One of the central ideas in medical oncology is that if you have two or more anticancer treatments that are active, you test them together to determine whether it's safe and whether the combination works better than each individually. We've been doing this with chemotherapy combinations for decades, but it's only been in the last few years that we have had more than one targeted therapy in lung cancer with enough activity to move ahead with combination work. Moreover, combinations with new agents is often limited by practical issues like companies needing to cooperative to provide their novel agents to each other if they aren't commercially available.

The combination of avastin (bevacizumab), an anti-angiogenic drug, with tarceva (erlotinib), an oral epidermal growth factor receptor (EGFR) inhibitor, has been among the best studied targeted therapy combinations. Both agents have been commercially available for several years in oncology, reducing one practical barrier. Secondly, it makes good sense to try to combat the cancer by treating both the cancer cell's growth and division signaling pathways (with tarceva) and the supporting microenvironment (reducing the tumor blood supply with avastin):

(Click to enlarge)

The fact that both drugs are marketed by Genentech meant that it was even more likely that we'd see them paired together.

But for the combination to really take off, it needs to offer some added value beyond just using them individually. The first hurdle was just showing that the combination was safe and had enough suggestion of clinical activity to continue to study the combination in larger trials. Several years ago, leading lung cancer researchers Alan Sandler at Vanderbilt and Roy Herbst at MD Anderson led a trial together that enrolled 40 patients who had previously received one line of therapy and didn't have any known contraindications to getting avastin (abstract here). They reported that there were no serious bleeding events or any other dreaded or unforeseen side effects, and the median survival of just over a year certainly justified pursuing further studies. One that did follow was a trial of 120 second line, avastin-eligible advanced NSCLC patients who were randomized to receive chemo alone (alimta or taxotere, physician's choice), the same chemo with avastin, or tarceva with avastin (abstract here). In this study, both the chemo/avastin and taxotere/avastin had a better response rate, progression-free survival, and six month survival than chemo alone. However, there was at least one fatal bleeding complication in each of the avastin arms, so it's not a risk-free treatment, and the results were not definitive with approximately 40 patients per arm.

But there are two big trials addressing the combination of avastin and tarceva that are particularly important because they are large enough to provide a real test of what the combination provides over one or the other alone. The first of these is called the BeTa (Bevacizumab/Tarceva) trial, which asks whether giving avastin along with second line tarceva improves survival compared with tarceva alone. This study enrolled about 600 patients with previously treated advanced NSCLC, and as the trial continued, they liberalized the enrollment to allow patients with previously treated, asymptomatic brain metastases or peripheral squamous lung cancers, far from the central blood vessels (most squamous cell NSCLC tumors are central, so there weren't many of these patients). A press release just reported that this trial was negative for the primary endpoint of overall survival, but the combination did apparently show an improvement in progression-free survival compared with tarceva alone. What this suggests is that there was some improvement in efficacy from the combination that was likely diluted by subsequent treatments that brought the tarceva arm closer to the survival of the combination arm over time.

We don't have any specific numbers, so this could be an impressive near miss, but we need more details. We also need to clarify whether the side effects were problematic in the combination arm, including whether the patients with treated brain mets or peripheral squamous cell tumors had any greater safety problems than the traditionally avastin-eligible patients.

The other major trial testing the combination of avastin and tarceva is called ATLAS, in which avastin-eligible patients started with four cycles of platinum-based doublet chemo (various combinations) with avastin, and then non-progressing patients were randomized to maintenance avastin alone or tarceva with the avastin (so, as the converse of the BeTa trial, is there a benefit to adding tarceva when avastin alone is the standard?). That trial of approximately 1150 patients was recently completed, I believe, so there's no dilemma about continuing the trial, but we don't have any additional information about the combination yet either.

But there are other trials studying this combination, including a couple of national phase II trials that I lead through SWOG. One is SWOG 0635, enrolling 80 patients with BAC, and the other is SWOG 0636, enrolling 80 never-smoking patients with advanced lung adenocarcinoma. Both of these trials are ongoing, but we'll need to have further discussions to see what changes should be made in the consent process, and potentially in the trials themselves, in light of the fact that the BeTa trial was negative for an overall survival benefit. It's certainly not a favorable development, but with a significant improvement in progression-free survival, it's not clear that the combination is a lost cause either.