Hacker News new | past | comments | ask | show | jobs | submit login

> It was a small trial, just 18 rectal cancer patients,

just 12

"All 12 patients (100%; 95% confidence interval, 74 to 100) had a clinical complete response, with no evidence of tumor on magnetic resonance imaging, 18F-fluorodeoxyglucose–positron-emission tomography, endoscopic evaluation, digital rectal examination, or biopsy. "

https://www.nejm.org/doi/full/10.1056/NEJMoa2201445




As with most experiments there's a control group (6 patients). The original statement of 18 participants is correct along with yours of 12 patient treatment group.


But then I guess it is slightly less surprising that all 12 people had remission, rather than all 18. (Or if all 18 had remission, then that's amazing but might not be to do with the drug.) Maybe it is still a great result, just slightly less significant.


You're correct. We'd expect to see more extreme outcomes with smaller sample sizes. The question is does this translate to more drugs when administered early which was the real reason for this study.


A placebo control group when there is an established standard of care? That sounds highly unethical. The above link to the abstract don't mention this and I don't have access to the full text paper. If you do can you clarify what the control arm received?


I intentionally avoided using placebo group. The control cohort may not have had cancer at all

_____________

The plan is to enroll six patients with MSI, regardless of their primary cancer diagnosis. This cohort will serve to generate hypothesis and initial data to plan a larger study. All analyses from this cohort will be exploratory

https://clinicaltrials.gov/ct2/show/NCT04165772


I appreciate your good intentions, but you should consider not writing things on the internet like this - as both of your posts are effectively misinformation.

> As with most experiments

This isn't true. It's absolutely standard in earlier trials of investigational agents to not have any sort of control arm.

> there's a control group (6 patients).

This isn't true. Within the study you posted, there are two different cohorts, with different patient types included.

> The control cohort may not have had cancer at all

This isn't true. The group you're calling a control cohort (cohort 2) must all absolutely have cancer, and are all actively treated with the study drug (TSR-042).


I appreciate your feedback. That said this is a message board for a bunch of nerds so hopefully people would discuss this with an oncologist.

>> As with most experiments

>This isn't true. It's absolutely standard in earlier trials of investigational agents to not have any sort of control arm.

Can we agree that Clinical trials are a small subset of all experiments? And that Phase 1 isn't as large?

For your second two points you're correct and I agree with you.


I’d encourage comment OP to continue writing comments online, so that you can learn. The damage done in misinformation here is probably extremely limited.

If comment OP was trying to misinform on purpose, it’s a different story as there’s no opportunity for improvement there.


Is this the same study?

https://www.healio.com/news/hematology-oncology/20220605/dos...

> At the time of presentation, 18 patients were enrolled on trial.

> Results among the 14 patients with at least 6 months follow-up showed a complete response among all patients (95% CI, 74-100), with no evidence of tumor on biopsy, digital rectal exam, endoscopic visualization, fluorodeoxyglucose-PET or MRI. The other four patients are responding to treatment.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: