No need to change post-operative treatment for lung cancer, says clinical scientist

Post-operative chemotherapy in non-small cell lung cancer (NSCLC) looks like an attractive proposition to prevent the cancer recurring, said Professor Nick Thatcher, of the Christie Hospital, Manchester, UK at the ECCO-12 conference in Copenhagen today (Tuesday 23 September). But the recent IALT study, which is said to demonstrate a survival advantage for patients given adjuvant treatment, was statistically unreliable and needed to be taken in context, he said.

“A large body of evidence – 17 cisplatin trials as opposed to 1, the IALT study -demonstrated no survival advantage with the use of cisplatin-based adjuvant chemotherapy”, said Professor Thatcher.

A meta-analysis of 14 trials of adjuvant (post resection cisplatin therapy) in NSCLC not only showed no survival benefit with chemotherapy, said Professor Thatcher, but with adjuvant radiotherapy (which was included in the IALT study) there was a survival detriment, 7% more people treated dying over 2 years. Three other more recent randomised cisplatin trials had also failed to show a survival benefit among patients receiving adjuvant chemotherapy in NSCLC, he said.

In the case of the IALT study, a probable survival benefit of 4.1% was suggested at 5 years in those who received adjuvant cisplatin compared with the no-chemotherapy control arm. “But there were only 1,867 patients in the trial, as against the intended 3300”, said Professor Thatcher. “This means that statistically the result is less strong than it should have been. Secondly more post-operative radiotherapy was given to patients in the control arm, possibly reducing the control survival disproportionately”.

“It is a shame when change of practice is undertaken on the basis of one unconfirmed result. A further trial or meta-analysis is ideally needed to determine the confidence of the result” said Professor Thatcher. “Although it is tempting to get excited about one positive result, it is vital that it should be properly validated by subsequent research. Furthermore, it is important to interpret trial results, not just in terms of ’p values’ but in terms of the confidence intervals which in the IALT trial fall within the range of all the other negative trials.”