Interested in @statsepi's take on the #pvalue# debate? Would you like to see the #OERs# from @Maths_MTU on #StatisticalAnalysisPlanning#? If yes, then check-out the #hybrid# event ⬇️⬇️⬇️. Registration - #RIMTU# @RI_MTU @gradstudiesmtu

On 5th November 2019 a pivotal trial of the kidney drug avacopan missed. The two sided p value was 0.1025. Four days later a ChemoCentryx executive emailed a colleague, “We cannot afford to miss a superiority outcome here.” The FDA set out the rest in a notice dated 27th April 2026. Company staff searched the unblinded data and picked nine patients whose scoring could be revisited, six on avacopan and three on placebo. On 13th November, before any re-scoring was requested, a company executive asked an outside statistical consultant to rerun the analysis on what the notice calls “a hypothetical dataset accounting for the anticipated results.” It assumed five of the six avacopan patients would flip to remission and none of the placebo three would. He replied that the new p value showed statistical significance. The re-scoring went out the next day. It came back exactly as the hypothetical had assumed. Five of six avacopan patients flipped. No placebo patient changed at week 52. The FDA notes that no week 52 change was requested for them. The database locked again on 20 November at p 0.0132. Approval followed in 2021. The New England Journal of Medicine retracted the paper on 29th June. The European Commission revoked the European licence on 4th August. Health Canada has told doctors not to start new patients. Britain and Australia are reviewing. TAVNEOS is still sold in America. Amgen bought ChemoCentryx in 2022, three years after the conduct described, strongly disagrees with the FDA, and has asked for a hearing. Second quarter sales were $150 million. Amgen carries $2.4 billion of TAVNEOS rights on its books, $2.3 billion of it American. At 30th June it had written off none of it.

In June a conference put Regeneron's bone-disease drug at a 99.9 percent reduction, with a p-value of .001. The label the FDA approved on Wednesday says that same measurement was not statistically significant. Both are true, and Pasatru works. The approval rests on a different measurement, the count of new bone growths, which the label reports as statistically significant at both doses. Nothing here says otherwise. The disease is fibrodysplasia ossificans progressiva. Muscle and tendon turn to bone, a second skeleton grows over the first, and about 900 people worldwide are diagnosed with it. Sixty three of them were in this trial, and there will never be a bigger one. The measurement that changed its mind is the volume of new bone. At week 56 the placebo group averaged 10.5 cubic centimetres of it. The groups on Pasatru averaged 0.02 and 0.01. Those three numbers are where the 99.8 and 99.9 percent come from. Under the test the trial specified in advance, that difference does not clear significance, and the label says so in one sentence. Under a different model, built after the trial had finished, it becomes 99.9 percent with a p-value of .001. That is the version presented in June and reported on the fourteenth. Neither the FDA's approval announcement nor Regeneron's press release mentions bone volume. The word does not appear in either document. There is a second number doing the same thing. The label permits dropping from 10 milligrams per kilogram to 3 if the higher dose is not tolerated. At 10 the flare-up rate ratio against placebo is 0.12, with a range of 0.03 to 0.42. At 3 it is 0.85, with a range running from a 74 percent fall to a 173 percent rise. Regeneron's release reports the 0.85 as a 15 percent reduction, which is correct, and prints no range beside it. That is exactly why the ranges matter here. When a disease has 900 known patients, the label is the only document that still carries how uncertain the numbers are. The conference slide does not. The press release does not. The coverage does not. By the time a number reaches a patient it has been through three retellings, and the range is gone from all of them.

$ALNY -5% this morning Sellside comments re: the 1.14 event rate ratio from the stabilizer (combo) arm of the CARDIO-TTRansform study: Paul Matteis, Stifel "While minimal benefit was largely expected for the eplontersen/tafamidis combo arm in CARDIO-TTRansform, it's an incremental surprise to see that the combination had a hazard ratio of 1.14--clearly worse than expected." Michael Leuchten, Jefferies: "The presentation showed a treatment effect of rate ratio (RR) 1.14 for the baseline stabiliser group vs overall RR of 0.89, with a strong enough interaction p-value (0.017) suggesting a potential differential treatment effect between patients without stabiliser at baseline. The primary endpoint curves in patients that were on stabilisers at baseline do not show an improvement over time, if anything there is a worsening rate in the combination group." Cory Kasimov, Evercore: "...the 1.14 RR in pts on baseline stabilizers was the key metric investors cared about most coming in and will likely not be viewed favorably as it relates to implications for ALNY. At first glance, the results appear less suggestive of a trial-design issue and more consistent with a potential limitation to the incremental benefit of dual therapy with silencers and stabilizers, further entrenching the competitive positioning of the stabilizer class."

I'm a cardiologist. Today the FDA approved something I've been waiting two decades for: the first oral PCSK9 inhibitor in history. A once-daily pill — Lipfendra (enlicitide), from Merck — that lowers LDL as powerfully as the injections we've relied on. This is one of those days that quietly changes the trajectory of the disease that kills more people than anything else on earth. Let me tell you exactly why — and who should act on it this week. Start with the biology, because it's elegant. Your liver cells are covered in LDL receptors — molecular catcher's mitts that grab "bad" cholesterol out of your blood and pull it in for disposal. The more receptors you have working, the lower your LDL. But your body makes a protein called PCSK9 whose entire job is to find those receptors and drag them to the incinerator before they can be reused. PCSK9 is the saboteur. Block PCSK9, and the receptors survive. They get recycled back to the surface. They keep clearing LDL, over and over. Cholesterol plummets. It is one of the most powerful mechanisms we have ever discovered in cardiology. And until today, you could only access it through a needle. That needle was a bigger problem than most people realize. The injectables — Repatha, Praluent — are phenomenal drugs. But needle aversion, injection-site reactions, pharmacy logistics, and cost meant that millions of patients who desperately needed them simply said no. I have watched patients stay at heart-attack-level cholesterol for years because they wouldn't inject. The barrier was never the science. It was human. Now the trial data — from a program spanning over 19,000 participants — and it's genuinely impressive. Roughly 60% LDL reduction. A placebo-adjusted 56% in the main lipids trial, 59% in patients with inherited high cholesterol. On top of statins. That matches the injectable class in a daily tablet. But here's the detail that made me sit up as a cardiologist — the part most coverage is burying: enlicitide also significantly lowered ApoB and Lp(a). If you follow me, you know why that matters. Lp(a) is the genetic, inherited cholesterol particle that roughly one in five people carry at dangerous levels — the one that triples heart attack risk and that diet, exercise, and even statins barely touch. We've had almost nothing oral that moves it. A daily pill that lowers LDL, ApoB, AND Lp(a) simultaneously is a genuinely meaningful expansion of what we can do. Who should pay attention today: Anyone above their LDL goal despite maximum statins or ezetimibe. The roughly 9.8 million high-risk Americans still not at target. Anyone with familial hypercholesterolemia — the genetic condition that loads dangerously high LDL from birth and causes heart attacks in the 40s and 50s. These patients have avoided the injectables for years. Statin-intolerant patients who need powerful additional lowering. Anyone who's had a heart attack or stroke and needs to drive LDL far lower than a statin alone can manage. The price is notable too: about $315 a month, meaningfully below the injectables — though it'll still require insurance prior authorization, and formulary placement isn't set yet, so access won't be instant. Now the honest caveats, because I'm a physician, not a hype man: This approval is based on LDL lowering, not yet on proven outcomes. The big cardiovascular outcomes trial — CORALreef Outcomes, over 14,500 patients — is still running. We expect the heart-attack reduction to follow the LDL drop, because the link between lower LDL and fewer events is one of the most rock-solid relationships in all of medicine. But the definitive proof isn't in yet, and I'll tell you that honestly. And a pill isn't automatically right for everyone. Some patients love a shot they take twice a month and never think about again. This is a powerful new tool, not a mandate. But make no mistake about the size of this. For twenty years I've told patients that LDL is one of the most important numbers they will ever own — and that for the right person, lower is better, and lower earlier is better still. Today the tool to get there stopped requiring a needle. The medicine to protect your heart just got radically easier to take. And easier medicine is medicine people actually use. If you're on a statin and still not at goal — or high cholesterol runs in your family and you've been dodging the shots — this is the week to call your doctor and ask one question: "Is the new oral PCSK9 inhibitor right for me?" The needle was the wall. Today the wall came down. Know your numbers. The science to change them keeps winning.