Why would Singapore authorize molnupiravir?

“If you drink much from a bottle marked 'poison' it is certain to disagree with you sooner or later.”

— Alice, somewhere in Wonderland.

One of the more baffling news stories last month was Singapore’s authorization of Merck’s molnupiravir. Yes, the same drug that is purported to work by introducing errors into COVID-19’s genetic code — which is, consequently, the most blatant example of the myopic tendencies of the pharmaceutical industrial complex and our public health officials. It almost goes without saying that the chances of introducing enough “errors” into the genetic code of a virus with a rapid rate of replication outside of a petri dish will do little or nothing for recipients while posing a high mutation risk.

In fact, the trial results showed that the drug provided little to no benefit. While some of the hysterical type were hopeful based on the interim results, a closer examination showed that the purported efficacy came from the randomization process. Prior to the trial’s interim analysis, the drug showed high efficacy with a placebo group that had a much higher rate of high risk individuals. Following the interim analysis, into the second half of the trial, randomization was better and the drug had strictly negative efficacy. In the end, the results were barely statistically significant in a shortened, and tiny, trial.

You can read more about my concerns with the randomization here:

One pill makes you larger and one pill makes you small. The ones that the Food and Drug Administration give us don't do anything at all.

Following the trial, letters to the editor ensued at several major journals. PhD’s galore were deeply troubled, concerned, or confused by the results. These letters make for excellent reading. Here are some excerpts:

“This difference was driven by an increased benefit with placebo in the post–interim analysis phase, with a primary outcome event occurring in 20 of 324 participants who received molnupiravir and in 15 of 322 participants who received placebo. The disparity between these periods is so large that the difference is statistically implausible.”

Kristian Thorlund (Ph.D.), Kyle Sheldrick (M.B., B.S.), and Edward Mills (Ph.D.)

Letter to NEJM

Honestly, at least they pointed out the obvious, but they did not finish their thought. They could have added “…if the drug works” for extra clarification.

“However, examination of Figure S3 in the Supplementary Appendix of their article (available with the full text of the article at NEJM.org) reveals a much simpler explanation: this drug is relatively ineffective against the B.1.617.2 (delta) variant”

Susan Levenstein (M.D.)

Letter to NEJM

Susan thought she found Occam’s razor because she aggregated the efficacy from both phases of the trial by variant type! Of course, she completely ignored the heterogeneous outcomes in the different phases of the trial, so she did not actually say anything that was of any value to anyone. She just repeated Merck’s press release in a roundabout way. Being “relatively ineffective” against delta does not explain negative efficacy in the second half of the trial.

“First, the estimates are very imprecise. The absolute risk reduction with respect to hospitalization or death through day 29 varies from 0.1 to 5.9 percentage points, which corresponds to a number needed to treat ranging from 17 to 1157. Second, although vaccination coverage against Covid-19 is high in most Western countries, the efficacy of molnupiravir among vaccinated persons is unknown, and a lower baseline risk of complications might result in an even higher number needed to treat. Third, obesity (in 74% of the trial participants) was the main risk factor for severe disease in this young population (median age, 43 years). The applicability of the results in adults presenting with other risk factors, including age greater than 65 years, remains uncertain. In addition, the person-time at risk with 1433 participants was too short to establish a relevant risk–benefit balance of a drug”

Dominique Roberfroid (Ph.D.), Vicky Jespers (Ph.D.) and Frank Hulstaert, (M.D.)

Letter to NEJM

I reproduced most of the above letter because it is simply a great take. Basically, everything is unknown and the trial proved nothing.

“When the trial was published, the authors reported a smaller absolute difference in primary outcome (−3%, 95% confidence interval −5.9 to −0.1, P=0.043) than quoted in Merck’s earlier press release (−6.8%, P=0.0012). As the upper limit of this 95% confidence interval approaches zero, these published results have borderline significance, indicating that even a small number of misclassified outcomes could overturn the significance of the findings. Robustness is further undermined by a modified intention-to-treat analysis that excluded 25 patients after randomisation. Whether a standard intention-to-treat analysis of all randomised patients would have produced significant results remains unclear.

The weakness of the evidence supporting molnupiravir has been partially obscured by the large relative benefit reported in Merck’s original press release. This created a cognitive bias by anchoring public and scientific opinion to a belief in substantial benefit. Press releases can also set unreasonable expectations among the public and policy makers, leading to demand for immediate access to poorly evaluated drugs.”

James Brophy (Ph.D., M.D.)

Letter to BMJ

I suggest reading James’ entire letter as he makes a few more excellent points both about the study itself and more broadly about the way drugs are approved.

Anyways, back to the topic of Singapore (I haven’t forgotten). Aside from the fact the molnupiravir trials were inconclusive, the trial participants were all unvaccinated. Singapore is vaccinated at an extremely high rate, thanks to dictatorial control over their citizenry, so they approved a drug based on results that are completely inconsequential to their target population based on variants that are now extinct. But even more strangely is the fact that there are two, much larger, ongoing trials for molnupiravir in vaccinated populations during the omicron wave that could have fully informed their results. Surely, health officials in Singapore knew these trials were being run, yet they couldn’t wait a few more weeks?

The PANORAMIC trial is being run with over 10,000 participants (with a separate, equally large trial being run for Paxlovid), and is purported to be completed around the end of May, and the RECOVERY trial with 8,000 participants began in late January.

What benefit does Singapore gain by rushing this approval a couple of months before such large trials are completed especially considering the lackluster initial results and high risk potential? What do they have to gain?