Post-operative, or adjuvant, chemotherapy is a standard approach for higher risk patients with resected early stage NSCLC, based on several randomized trials that have been presented and published in the last few years that show a survival benefit from chemotherapy. All of the trials that have shown a statistically significant survival benefit have given chemotherapy after surgery, but it’s hard to envision why the same chemotherapy given before surgery wouldn’t be just as good or better. Pre-operative, or neoadjuvant, chemo allows treatment of potential micrometastases at the earliest possible opportunity. Probably more importantly, if it’s helpful to give chemotherapy to reduce the risk of recurrence with early stage NSCLC, we’d suspect that more patients can actually get their intended chemo if it’s given before surgery vs. after.

Still, we don’t have the trials to support the idea that neoadjuvant chemotherapy is as good or better than adjuvant chemotherapy. Some trials of pre-operative chemo have suggested a similar magnitude of benefit as post-operative chemo, but haven’t been large enough to have the improved survival be significant. Overall, we might suspect that neoadjuvant chemotherapy is a strong alternative, and a “meta-analysis” of multiple studies showed that three was no significant difference in outcomes between these strategies, but neoadjuvant chemotherapy hasn’t been studied well enough to be readily adopted as a standard of care.

At this year’s ASCO meeting, Dr. Enriqueta Felip from Spain presented the results of a trial that we hoped would provide an opportunity to really test these two alternatives directly. The NATCH (Neoadjuvant vs. Adjuvant Carbo Taxol Hope) study randomized 624 patients with clinical stage I – III (staging based on appearance of scans) NSCLC to pre-operative chemotherapy with three cycles of carboplatin/taxol (paclitaxel), the same chemo given post-operatively, or surgery with no chemo before or after.

NATCH Trial Schema

At the end of the day, the main findings were that pre-operative chemo was associated with the same survival as post-operative chemo, so we say comparable results, but neither of these chemo arms had a better survival than surgery alone.

How could we have seen such disappointing results when our recent results have consistently demonstrated a benefit from chemotherapy?

In the end, I think we learned from this trial, but not what exactly what we hoped to learn. One confirmatory lesson was that more patients did receive the chemo that was intended if they did it pre-operatively. Of those assigned to the neoadjuvant arm, 96% received chemo and surgery, and 90% received all three cycles along with surgery. In contrast, only 66% of the patients on the adjuvant chemotherapy arm received any chemotherapy, and only 61% received only three cycles. So it’s important to recognize that the patients on adjuvant chemotherapy trials aren’t the same as the ones on the neoadjuvant chemotherapy trials, since we’ve winnowed out 1/3 of patients who have had complications or just don’t feel up to more treatment.

The NATCH trial also highlights the importance of looking at the patients actually likely to benefit from chemotherapy along with surgery. The risk of recurrence is higher in patients with higher stage cancer, and the value of chemotherapy is tenuous and perhaps even harmful for patients with stage I NSCLC. But which patients were enrolled on the NATCH trial? Three quarters had stage I NSCLC, the group for which it’s hardest to show benefit, and about 10% actually had stage IA tumors, for which evidence strongly suggests that chemotherapy is detrimental.

In fact, the trial showed that there was no hint of a difference in disease-free survival (DFS) at 5 years among patients with stage I NSCLC, but both arms showed a better 5-year DFS with chemo, particularly the pre-operative arm.

DFS by Stage on NATCH

The trial was also squeezed on the other end, namely that there were also many patients who had more advanced stage disease than the trial was meant to study. The study enrolled patients based on clinical staging (which didn’t include PET scans for all, or I believe even most, patients), and when they actually did surgery, they found that about ¼ of the patients on the study actually had mediastinal node involvement, or even metastatic disease.

So even though there was no meaningful difference in survival among the three arms, we see a suggestion that the group actually most likely to benefit from chemotherapy very likely does. But when a trial with about 200 patients per arm has the majority of patients with stage I NSCLC on one side and locally advanced or metastatic disease on the other, it really doesn’t have a chance to test what it was designed to test. Another question is whether a cisplatin-based chemotherapy regimen would have shown a stronger benefit, since carbo-based chemo hasn’t yet been shown to improve survival for patients with resected NSCLC.

There is an Asian study also looking at the question of neoadjuvant vs. adjuvant chemotherapy, so we may someday get a more meaningful answer. For now, however, post-operative chemotherapy remains better studied, but there is a growing amount of evidence that the pre-op alternative is reasonable and may even ultimately emerge as a stronger choice, if only we can test it properly.