For many patients with early stage, resected NSCLC, chemotherapy after surgery may be a strong consideration to minimize the chance of the cancer returning, in which cases, it is often not possible to cure it. Several clinical trials over the past few years have shown benefits from chemo combinations, but which ones would be the leading considerations now?

The first trial that showed a significant benefit of chemo after surgery, the International Adjuvant Lung Trial (IALT abstract), gave any of four chemo regimens, each combining cisplatin with another drug. The most commonly used were etoposide and vinorelbine (also known as navelbine), with the other two choices rarely used in lung cancer these days, at least in the US. Two other trials, one run through the National Cancer Institute of Canada (abstract here), and another done in Italy and presented at international meetings but not yet published, also used cisplatin and navelbine as the only combination used and had better results. So there’s a lot of evidence in favor of cisplatin/navelbine, and it’s a very reasonable and arguably among the best choices in this setting, certainly if you follow the data strictly.

From the setting of metastatic NSCLC, however, we know that the platinum-based two drug combinations that are commonly used are all nearly identical in how effective they are. Cisplatin and navelbine has been compared to other doublets and has been in the same range. One trial in advanced NSCLC (known as TAX 326, abstract here) actually indicated that cisplatin and docetaxel (taxotere) was actually a bit more effective, with a slightly longer overall survival in patients with metastatic disease, than cisplatin/navelbine. That’s a rare case where one doublet outperformed another, but by and large they are remarkably similar in activity and primarily differ in schedule and side effects. So cisplatin/navelbine, cisplatin/taxotere, and cisplatin/gemcitabine (also known as gemzar) would all be very strong considerations. Cisplatin and paclitaxel (taxol) could also be done, but most oncologists don’t tend to use that combination.

The bigger question is whether substituting carboplatin for cisplatin is appropriate. Carboplatin-based doublets generally come out in the same ballpark as cisplatin doublets, and carboplatin is usually considerably better tolerated, so most oncologists favor that in the metastatic setting, where toxicity of treatment is especially important because we can’t realistically plan to cure patients. However, I and many other lung cancer experts believe that cisplatin is slightly superior in activity, and therefore favor using it over carboplatin in the curative setting, despite the greater likelihood of side effects. A few clinical trials and also a recent meta-analysis (cisplatin vs. carboplatin meta-analysis abstract) indicate that cisplatin is associated with modestly better survival in the metastatic setting, and this combined with the fact that all of the data showing a significant survival benefit after surgery used cisplatin makes me favor cisplatin combinations if patients can tolerate it.

There was a very important trial, known as CALGB 9633, that used carboplatin and taxol, after surgery. This was a smaller trial, with 344 patients, less than they had planned, and it included only patients with stage IB disease, who don’t look like they get the same degree of benefit from post-op chemo as stage II or IIIA patients. In the CALGB 9633 trial, the preliminary results (preliminary CALGB abstract) showed a significant benefit vs. no chemo, so most of the lung cancer community felt that chemo with either a cisplatin or carboplatin doublet was fine. However, the updated results (which have only been published as an abstract thus far) showed that the benefits were much less dramatic with longer follow-up. It is possible to consider the CALGB trial as negative or possibly just not positive enough because it was too small to show a significant difference in a group of patients with a good prognosis already. But at the end of the day, the data are not as strong for carboplatin, and the existing evidence suggests that cisplatin is a bit more effective, so that is why I and most other experts in the lung cancer community prefer cisplatin in the adjuvant setting, at least in the patients who are well enough to pursue that approach.