Thread by @Garth_Nicholas1

22 November: PACIFIC, intention to treat

For Lung Cancer Awareness Month #LCAM I’m going to summarize 30 important lung cancer trials over 30 days. These posts are directed at non-medical professionals, with descriptions of the results and of what makes a good trial. #lcsm 1/13

We previously looked at locally-advanced lung cancer on 4, 6, and 9 November. We have established standard treatment as ~60 Gy radiotherapy with concurrent chemotherapy for those that are not resectable by lobectomy. Today’s trial looked at adding immunotherapy. 2/13

The antibody in this trial is durvalumab. Like the previously mentioned pembrolizumab (Nov 13, 16) and nivolumab (Nov 19), durvalumab inhibits the interaction of PD-1 and PD-L1. Unlike the other two, durvalumab binds to PD-L1. Clinically, the difference seems negligible. 3/13

This trial enrolled people who had completed chemoradiation for locally advanced lung cancer. They had to have a CT scan showing that their cancer had not worsened despite chemorads. Then they were randomized to either 1 year of durvalumab or placebo. 4/13

Progressions-free and overall survival were co-primary endpoints, tested at p=0.025, with the trial being positive if either end point was met. The power was 95% to detect a PFS HR of 0.67, and 85% for an OS HR of 0.73. (see 12 and 13 Nov for power and HRs) 5/13

Do you remember my thread from 17 November when I mentioned the skepticism that might greet a hypothetical analysis of 40 subgroups? I’m not saying; I’m just saying...
6/13

The study demonstrated substantial improvements in both progression free and overall survival. Longer follow-up will be required (see our discussion of the right side of the curve, 7/13

It’s interesting to compare the standard arm in this study to prior studies in stage III disease. In a 1999 study establishing concurrent chemorads as the standard, median OS was 16.5 months, and 2-year survival 34.6%.
In this study, they’re 28.7 months and 55.6%, respectively.

These improvements are probably attributable to:

1. Better staging (particularly PET scans)
2. This trial’s requirement for post-radiation CT scan
3. Improved radiation techniques (see 9 November)
9/13

Most of our trials have indicated that they analyze patients based on “Intention-to-treat”. What does this cryptic phrase mean?

It means that patients are analyzed in the group they were randomized to, regardless of whether they actually receive any or all of the treatment.10/13

You might think that if a patient assigned to a year of immunotherapy dropped out after one dose that it might be fair not to include them in the anlaysis: they didn’t really receive the study treatment after all, and aren’t a fair measure of whether or not the treatment “works”.

Such a practice would preferentially exclude less fit people from the experimental arm, biasing the results in its favour.
Also, we want to know whether the drug works in the whole population: if 75% of participants have to drop out after one dose, that’s part of the answer.12/13

Tomorrow we’ll return to the world of ALK-positive lung cancer, and consider the pressing question of whether Rene Descartes can help us in understanding trials.
13/13