So far, I've only written a few introductory posts on mesothelioma, but there were some interesting presentations at ASCO 2007 about the topic. One described the results of an expanded access protocol (EAP), which is when a company gives free access to a drug that is not yet commercially available (generally in exchange for participation in a data-collection study). This EAP was for Alimta (pemetrexed) as a treatment for malignant pleural mesothelioma (MPM) in Europe (abstract here). On this EAP, patients with mesothelioma from 13 different European countries were enrolled and could receive cisplatin/alimta (the combination approved by the FDA after being studied in a large randomized trial (abstract here)), carboplatin/alimta (carbo being a popular substitution for cisplatin in many settings because it is associated with considerably less nausea, risk of kidney damage, hearing loss, and some other side effects), or alimta alone. All patients received B12 and folate supplementation with their chemo. Over 3000 patients with MPM had data collected on how they tolerated treatment and how well they did on therapy, with just over 2000 of them receiving this treatment as their first chemotherapy. The report described the experiences of the 1704 patients who received alimta with either cisplatin or carboplatin, rather than alimta alone, and compared the results. The groups were almost evenly split between the two combinations (843 getting cis/alimta, 861 getting carbo/alimta). The two groups were quite comparable except that the median age of patients getting carboplatin was about 66, vs. 62 for recipients of cisplatin. This isn't especially surprising, because the younger, presumably a little healthier patients would likely be selected to receive the more rigorous treatment.

The results showed that the two groups performed remarkably similarly in just about every way. The median number of cycles was 5 with cis/alimta vs. 6 with carboplatin/alimta, but otherwise, both groups had the same general response rate in the 25% range, the same disease control rate (stable or responding) of 75-80%, median time to progression of 7 months, and one-year survival rate of 60-65%. Looking at the numbers and the survival curves, it's impressive to see how overlapping they really were:

(Click to enlarge)

In fact, it's interesting to compare these results to the international randomized trial that led to FDA approval of cisplatin/alimta (abstract here), which is a very large trial, including 456 patients. But the EAP includes more than 3 times as many treated with a doublet of platinum and alimta in the first-line setting, and it's a "real world" experience of how regular patients in various cancer centers all throughout Europe did. In many ways, this is more valuable to the average patient than the results of a trial conducted at a much more limited set of treatment locations. And while the response rate with cis/alimta was higher in the randomized trial, at just about 41%, it's well known that response is notoriously hard to reliably assess in MPM, because you don't tend to have well circumscribed lesions to measure. The very important endpoints of progression-free and overall survival looked better in the EAP than in the randomized trial. While the median survival on the EAP wasn't specifically reported, the one-year survival corresponds to a median of more than a year. So the most important endpoint, survival, looks every bit as good or better in thevery large "real world experience" as it did in the randomized trial, and patients who received carboplatin/alimta appear to do just as well as those who received cisplatin/alimta. These results aren't quite as conclusive as having a randomized trial, but they go a long way to reassuring me that if we substitute carboplatin for cisplatin with alimta for MPM, whether due to a worry about toxicity or patient preference, it isn't likely to short-change that patient at all in terms of results. Good to know, because carboplatin/alimta is often quite well tolerated, with side effects primarily of decreased blood counts.

Of course, these agents are also used in other aspects of lung cancer, most notably advanced NSCLC. One trial presented at ASCO by Gronberg and colleagues (abstract here) randomized 437 previously untreated patients with advanced NSCLC to receive either carbo/gem or carbo/alimta for up to 4 cycles or progression of disease (then move on to later treatment if appropriate). The two arms had essentially identical survival and also quality of life assessments by patients on the study. However, patients who received carbo/alimta had significantly less severe drops in blood counts, including red blood cells (anemia), white blood cells (infectious risk), and platelets (associated with bleeding risk). These patients also received significantly fewer transfusions of red blood cells and platelets. Overall, while I've already said that the platinum-based doublets generally achieve very similar results, we're always looking for choices with greater convenience and fewer side effects. I use a good bit of carbo/gemcitabine because the main side effects are lab-based, and that usually means that I'm telling patients about side effects rather than them telling me (and they can't feel these "paper toxicities" usually). So if carbo/alimta as a doublet provides as good activity and lower rates of significant drops in blood counts, few other side effects, and it can be given just one day every three weeks, it's an attractive choice. Dr. Tom Lynch from Massachusetts General Hospital, who provided the commentary after Dr. Gronberg's presentation of the results at ASCO, also noted that this combination is a particularly attractive one.

Alimta is not approved by the FDA, however, for use in previously untreated patients with NSCLC, and at my own center I'm not able to give it to people who don't have MPM or previously treated NSCLC. But there are more studies coming out with alimta in the first-line setting, and others also looking at it in other lung cancer scenarios. At the moment, I'm limiting my use of it to where it's indicated, but given the nice balance of how tolerable it often (not always) is vs. its activity, I'd use it more if I had the opportunity. The only problem, though, is that we don't have a lot of drugs that have shown value in previously treated patients with NSCLC, and if I use alimta first-line, it's one less drug available for later. This is an issue with Taxotere already, since it's approved as a first-line or second-line therapy.

In the meantime, in the right situation, platinum-based combinations with alimta can be quite helpful, and the mesothelioma EAP from Europe provides evidence from a large group of patients that carbo/alimta can provide very comparable results to the generally more challenging cisplatin/alimta combination.