Some interesting work coming out of Denmark is evaluating the association of tumor ERCC1 expression with outcomes of inoperable patients receiving carboplatin and gemcitabine, and suggesting that the association of different outcomes depending on tumor ERCC1 expression is sex-specific.

To provide a very brief review of a subject covered in a prior post, ERCC1 is a protein that works to repair DNA damage, whether it is caused by a cancer or by certain chemotherapy drugs (which work by damaging the DNA of cancer cells), such as cisplatin or carboplatin. ERCC1 has been studied in a variety of cancers, and essentially the evidence has been pretty consistent in showing that patients with high ERCC1 levels seem to do better if not getting chemo (presumably because ERCC1 can help reverse the cancer's effect on DNA of normal cells) but do worse if getting chemotherapy, at least platinum-based chemo (because ERCC1 undoes the DNA-damaging effects of chemo on cancer cells).

The work out of Denmark looked at the association of survival with ERCC1 protein expression among 163 patients with inoperable NSCLC. In some parts of Europe they treat all inoperable people with chemotherapy rather than give chemo and radiation concurrently to potentially curable patients locally advanced or "medically inoperable" (not fit enough to tolerate surgery) disease, so this mix of patients is not just those with metastatic NSCLC. As has been seen in some other work with advanced NSCLC that showed better survival in patients with low tumor ERCC1 expression who received cisplatin-based chemotherapy, this trial showed an overall superior survival in patients who had low ERCC1 protein expression (58% of the whole population) and received carboplatin/gemcitabine. However, the difference in survival by ERCC1 status was only seem among the half of patients on the trial who were men. No difference in survival was seen among women with high vs. low ERCC1 expression.

This work is interesting, though with just 163 patients who have a range of stages being treated, the relatively small population and wide variability of patients limits what we can conclude. I think it's helpful to get some information showing a correlation of ERCC1 expression with outcomes on a carboplatin-based chemotherapy regimen, since we very commonly use carboplatin instead of cisplatin for patients with advanced NSCLC. The fact that the results showed a difference only in male patients is unexpected, since this hasn't been observed before, though a sex-based difference also hasn't been looked for in much of the past work on ERCC1. We are appreciating more and more that there may be important differences between men and women with lung cancer (besides the obvious ones), but we still don't know as much as we'd like to about that.

In the meantime, I would not presume that this is a clinically significant difference unless it's observed in other studies, since it could just be a random, spurious observation in a single relatively small trial. It does give us reason to pause about becoming too convinced that we know the whole story with ERCC1 and should be using it now for clinical decision-making. Though I have been impressed with the early work on it, I also recognize that this work has nearly all been retrospective, and we don't have any evidence yet that using it to prospectively assign treatment improves survival. For now, I see this work as one more piece of a puzzle that is still taking shape. Though we may be moving toward molecularly-based treatments, we need to ensure that we really understand the signals before we change our clinical practice to follow the early leads.