乌鸦传媒

Final efficacy and, perhaps more importantly, safety data on Pfizer's oral treatment for mild-to-moderate COVID-19 have finally appeared in peer-reviewed print, nearly 2 months after the FDA allowed the company to market the product.

In the phase II/III trial , involving 2,246 unvaccinated, non-hospitalized patients considered to be at high risk for COVID worsening, those assigned to the combination of nirmatrelvir and ritonavir (Paxlovid) were 88.9% less likely to die or be hospitalized compared with those on placebo, reported Pfizer researchers led by Jennifer Hammond, PhD, (NEJM).

Both groups were nearly equal in size, at about 1,040 each in the modified intention-to-treat analysis, which excluded about 200 patients who hadn't exactly met the prespecified inclusion criteria. Eight patients receiving the drug combination were hospitalized and none died; in the placebo group, 66 either were hospitalized or died. All 13 deaths in the study were in patients assigned to placebo.

The 88.9% efficacy figure was virtually identical to that from an interim analysis of 774 patients, which yielded an estimate of 89.1%. Pfizer had in November; the FDA put the figure at 88% in its -- appearing almost at the end of the 29-page document (the equivalent of an official label for authorized but not formally approved products).

The NEJM paper also fleshed out the safety data, which were relatively sparse in both Pfizer's interim topline report and the FDA fact sheet. Overall, in the final analysis, 22.6% of patients receiving nirmatrelvir-ritonavir developed adverse events during the trial compared with 23.9% of the placebo group. Serious adverse events were recorded for 1.6% of the active-drug group versus 6.6% of those on placebo; progressive COVID symptoms seemed to account for most of this difference.

New in the published data was information on adverse events considered treatment-related. Rates of these, not surprisingly, were higher with nirmatrelvir-ritonavir: 7.8% versus 3.8%. However, those with grades of 3 or 4 were not more common, with both groups recording five such events. Discontinuations or dose reductions due to events judged to be treatment-related were also about the same (11 vs 10).

A supplementary appendix listed what looked like every single adverse event, organized by organ and system. Overall, these supported the information contained in the FDA fact sheet, which stated that dysgeusia was the only drug side effect seen in more than 5% of patients and more commonly than with placebo (5.6% vs less than 0.3%). Others less common, but more so with the active treatment, included diarrhea, hypertension, and myalgia.

Also new in the NEJM paper were data on viral loads with treatment. At day 5 of treatment, these were lower by 0.868 log₁₀ with nirmatrelvir-ritonavir versus placebo, i.e., just short of tenfold, when treatment began no more than 3 days after symptom onset.

Patients enrolled in EPIC-HR were symptomatic, unvaccinated adults testing positive for SARS-CoV-2 infection and showing signs of COVID-19 illness no more than 5 days prior to randomization. They also had to have at least one known risk factor for severe COVID. Randomization was 1:1 to placebo or the nirmatrelvir-ritonavir combination. Individuals with severe liver or renal impairment or non-COVID systemic infections were excluded.

The prescribed treatment course was 5 days, which is also the duration specified by the FDA. Dosages were 300 mg for nirmatrelvir and 100 mg for ritonavir; each is packaged as separate pills to be taken together twice daily.

Nirmatrelvir was designed specifically to treat COVID-19; it blocks a SARS-CoV-2 protease critical to the virus's replication cycle. The point of including ritonavir is not for its own antiviral effect, but rather to inhibit nirmatrelvir's metabolism via the CYP3A enzyme, thus prolonging its lifetime in circulation. Because of this, the product shouldn't be used alongside other drugs that induce CYP3A or rely on its normal activity; indeed, the FDA fact sheet includes a list of such drugs that goes on for several pages.

Another trial called began last September that is testing the product for post-exposure prophylaxis, enrolling a projected 2,634 people with an infected household member. No results have been reported yet.

Comment