https://events-program.easl.eu/ilc2022/en-GB/...act/231073
I wanted to make sure everyone knew what p value meant. The idea is to have a very low p value. The very maximum the p value should be to be meaningful to FDA is 0.05. Anything higher does not cut it with the FDA, (for the most part). So the aim is to achieve a p value of less than 0.05.
What does p value mean? Essentially, it is asking: What is the probability that the results you achieved in your arm could have randomly occurred in the placebo group?
So a p value of 0.05 would be a 5% or 1 in 20 chance possibility that the results achieved could have randomly occured in the placebo, untreated group.
Looking only at the 350mg arm, in a dataset of only n=72 patients, we reached astounding p values: if this were a massive trial with 500 or more patients, the p values would have been far lower.
p = 0.008 or 1 in 125 chance that result could have randomly occurred in placebo group
p = 0.021 or 1 in 48 chance that result could have randomly occurred in placebo group
p = 0.020 or 1 in 50 chance that result could have randomly occurred in placebo group
p = 0.027 or 1 in 37 chance that result could have randomly occurred in placebo group
p = 0.009 or 1 in 111 chance that result could have randomly occurred in placebo group
the p value for the pooled group, 350 + 700 were:
p = 0.014 or 1 in 71 chance that result could have randomly occurred in placebo group
p = 0.013 or 1 in 77 chance that result could have randomly occurred in placebo group
the p values for the 700mg haplotypes were:
p = 0.006 or 1 in 167 chance that result could have randomly occurred in placebo group
p = 0.013 or 1 in 77 chance that result could have randomly occurred in placebo group.
Conclusions were understated. In this trial we wanted to know if there was a slim possibility that LL would work in NASH. We did not expect to knock it out of the park.
Partnership shall get 'er done.
Quote:Results: Analyses were conducted on the full analysis set (n = 72) of whom 44% were of Hispanic or
Latino ethnicity and 58% with baseline moderate to severe fibro-inflammation (cT1 875 ms). Mean
percent change from baseline PDFF was significantly reduced in the 350 mg group vs placebo (-5.94%
vs +9.85%, p = 0.008 ) but not in the 700 mg group (+3.75% vs +9.85%, p = 0.135 ). Mean change cT1
was significantly reduced in the 350 mg group vs placebo (-24.38 ms vs +27.64 ms, p = 0.021 ) but not
in the 700 mg group (-2.73 ms vs +27.64 ms, p = 0.059 ). Significant reductions were seen in the 350
mg subgroup with baseline cT1 875 ms in both PDFF and cT1 vs placebo (-4.37% vs +9.85%, p =
0.020 and -42.00 ms vs +27.64 ms, p = 0.011 ) respectively. In subjects with cT1 950 ms at baseline,
PDFF and cT1 were significantly reduced with 350 mg vs placebo (-9.39% vs +9.85%, p = 0.027 and -
68.85 ms vs +27.64, p = 0.009 ) respectively. Mean change in baseline to week 14 for M65 ELISA
(cK18 and K18) decreased in the 350 mg group (340.55 to 332.4 U/L; -8.18) while increased in
placebo (301.96 to 411.64 U/L; +109.78). In post hoc analyses, mean percent PDFF and mean cT1
were significantly reduced in the pooled 350 + 700 mg group compared to placebo (-1.09% vs
+9.85%, p = 0.014 and -13.30 ms vs +27.64 ms, p = 0.013 ) and in the 700 mg group with genetic
haplotypes known to over produce CCR5 compared to placebo (-27.9% vs +9.85%, p = 0.006 and -
45.4 ms vs +27.64 ms, p = 0.013 ). There was no grade 3 or higher drug related treatment emergent
adverse event. Injection site reaction and mild diarrhea occurred more frequently with leronlimab than
placebo but were not associated with discontinuation.
Conclusions: The primary endpoint (PDFF) and secondary endpoints (cT1) were met for the 350 mg
group and moderate to severe fibro-inflammation by cT1 350 mg subgroup at baseline. The pooled
350 + 700 mg group also had significant reductions in PDFF, cT1 vs placebo. Treatment with
leronlimab was well tolerated. These results are supportive of further trials with leronlimab for NASH.
I wanted to make sure everyone knew what p value meant. The idea is to have a very low p value. The very maximum the p value should be to be meaningful to FDA is 0.05. Anything higher does not cut it with the FDA, (for the most part). So the aim is to achieve a p value of less than 0.05.
What does p value mean? Essentially, it is asking: What is the probability that the results you achieved in your arm could have randomly occurred in the placebo group?
So a p value of 0.05 would be a 5% or 1 in 20 chance possibility that the results achieved could have randomly occured in the placebo, untreated group.
Looking only at the 350mg arm, in a dataset of only n=72 patients, we reached astounding p values: if this were a massive trial with 500 or more patients, the p values would have been far lower.
p = 0.008 or 1 in 125 chance that result could have randomly occurred in placebo group
p = 0.021 or 1 in 48 chance that result could have randomly occurred in placebo group
p = 0.020 or 1 in 50 chance that result could have randomly occurred in placebo group
p = 0.027 or 1 in 37 chance that result could have randomly occurred in placebo group
p = 0.009 or 1 in 111 chance that result could have randomly occurred in placebo group
the p value for the pooled group, 350 + 700 were:
p = 0.014 or 1 in 71 chance that result could have randomly occurred in placebo group
p = 0.013 or 1 in 77 chance that result could have randomly occurred in placebo group
the p values for the 700mg haplotypes were:
p = 0.006 or 1 in 167 chance that result could have randomly occurred in placebo group
p = 0.013 or 1 in 77 chance that result could have randomly occurred in placebo group.
Conclusions were understated. In this trial we wanted to know if there was a slim possibility that LL would work in NASH. We did not expect to knock it out of the park.
Partnership shall get 'er done.
19
