I recently received a question on the Q&A Forum about the use of cisplatin vs. carboplatin in SCLC. In contrast to the smoldering debate about cisplatin vs. carboplatin in NSCLC that I described in a recent post, there's been very little study and not as much debate about SCLC. What little I can say is that there was a trial published in 1994 by Skarlos and colleagues from Greece that randomized patients to cisplatin or carboplatin with etoposide, and it included patients with both limited and extensive SCLC (abstract here):

(Click to enlarge)

This trial is so old that the carboplatin dose is in mg/m2, which is an older way of dosing this agent than our current one that uses age, patient sex, and kidney function to calculate a drug exposure known as area under the curve, or AUC. Our current dosing of carboplatin is almost always by AUC. And just FYI, the dose of etoposide listed on my figure is different from the one listed in the abstract. I believe the latter is a typo/misprint, but I don't have the original paper availabel to double check that. The dose of 300 mg/m2 would be a staggeringly high dose, at least if we're talking IV and not oral etoposide.

The study demonstrated relarkably similar results for the two arms, with very impressive complete response rates of 57% and 58% and median survivals of 12.5 and 11.8 months for cisplatin and carboplatin, respectively. The toxicity profile favored carboplatin, with significantly lower rates of severe nausea/vomiting, neurotoxicity (neuropathy), and leukopenia (low white blood cell counts), and infections in the patients who received carboplatin. So this work suggests comparable results and perhaps a more favorable therapeutic index (balance of activity vs. side effects) for carboplatin.

That's one reason I modestly favor carboplatin over cisplatin outside of a clinical trial in ED-SCLC. Many additional trials have been done with carboplatin-based chemo in ED-SCLC that show survival about what you'd expect for cisplatin as well, and I'll soon give some additional information on one trial presented at ASCO 2007 that compared two carboplatin-based doublet regimens.

However, in a curative setting like LD-SCLC, I still favor cisplatin. There are no places in oncology where cure rates are better with carboplatin than cisplatin, but there are a number where cisplatin is signficantly (if modestly) better than carboplatin. In fact, the full paper showed some trends that looked a little worse for carboplatin recipients LD-SCLC, but the trial is small, and the differences weren't enough to be statistically significant. Given the high stakes and the much more extensive body of evidence with cisplatin in LD-SCLC (the best data we have in long-term cure rates in LD-SCLC are with cisplatin/etoposide and twice daily RT, as described in a prior post), that's my standard approach unless there's a reason a patient can't tolerate cisplatin. At least we have some evidence to suggest that patients who get carboplatin instead of cisplatin aren't compromising significantly and may do just as well, based on this limited Greek experience.