With last week's FDA approval of alimta in the first line setting for NSCLC, we're likely to see a lot of alimta (pemetrexed) use shift from the second and third line setting to first line. Alimta's been a very popular choice for previously treated patients, based on issues like the relatively convenient schedule of a ten minute infusion one day every three weeks, no hair loss, and typically less of a drop in blood counts than seen with some other regimens. In the last year, we've seen a growing amount of data using it in combination with either cisplatin or carboplatin in the first line setting. While the official approval of alimta was with cisplatin and only in patients with non-squamous cancers (see prior post), I suspect that it'll be the carbo/alimta combination that really makes an impact. Historically, US-based oncologists and their patients have preferred carboplatin for it's easier side effect profile and greater convenience (cisplatin is usually given with loads of fluids to protect the kidneys, leading to long days in the outpatient infusion center or an overnight hospitalization stay). And based on the responses I saw on a recent audience response system for a case-based lung cancer meeting this weekend, the greater interest will again be with carboplatin instead of cisplatin.

So what data do we have to support this combination in lung cancer? Early safety and efficacy studies were done in Italy (article here) and at MD Anderson Cancer Center in Houston (article here) and were certainly encouraging. A friend of mine at MD Anderson, Dr. Ralph Zinner, led their trial and raved about how well tolerated the regimen was, with some patients continuing without progression for 8 or 10 or even more cycles. With a median overall survival of over 13 months, these results were favorable (we expect most trials from MD Anderson to exceed the numbers seen for trials around the country, due to selection bias in the population who come there), but a special appeal was the tolerability of the regimen.

At ASCO 2007, a Norwegian phase III trial was presented that compared carbo/alimta to carbo/gemcitabine (abstract here). Not surprisingly, the two doublets had a remarkably similar median overall survival, and they also had a pretty much identical quality of life over the course of treatment.

In terms of side effect profile, carbo/alimta was a modest winner, based on less significant drops in blood counts and need for transfusion support, but I'd consider the carbo/gemcitabine regimen to also be among the quite well tolerated (mostly drops in blood counts, compared with nausea or hair loss or other directly experienced side effects). Here's the basically superimposed survival curves:

Frankly, a median survival of just over 7 months is very average, certainly not exceptional. They did include patients with a marginal performance status, who accounted for nearly a quarter of the study population, so this may have contributed, since most of the historical trials included few or no patients with a performance status of two. The other issue is that this study was conducted in Europe, where second and later line treatments are not routinely administered, so patients may have done a little worse than they would have if more had received subsequent treatment.

One other issue in current clinical decision-making is that avastin is standardly combined with chemo for the subset of patients who are eligible to receive avastin. While the decisive trial that led to the approval of avastin used carbo/taxol (abstract here), many oncologists are inclined to add it to whatever chemo regimen they'd prefer to use. We received some actual evidence on the safety and efficacy of a combination of carbo/alimta/avastin in a phase II clinical trial that was presented at ASCO this year (abstract here). My friend Jyoti Patel at Northwestern led this trial of 51 patients who received 6 cycles of carbo/alimta/avastin, then switched to alimta and avastin if they hadn't progressed:

With a median survival over a year, that's very encouraging, although it's not a large enough study to presume those numbers would be achieved in a broader population.

To me, one of the main concerns I'd have about using alimta first line is that it's an effective and quite tolerable second or third line option, when the list of good choices is short. Perhaps transitioning from doublet to single agent alimta or alimta/avastin after 4-6 cycles of platinum-based chemo is an attractive option -- several new trials are building this into their treatment plan. In the meantime, cisplatin or carbo with alimta is likely to be an increasingly used option, although the evidence would support only using it in the 70-75% of patients who have non-squamous NSCLC.