A randomized, double-blind, placebo-controlled trial of IL-7 in critically ill patients with COVID-19

Abstract

Background: Lymphopenia and failure of lymphocytes to mount an early IFN-γ response correlate with increased mortality in COVID-19. Given the essential role of CD4 helper and CD8 cytotoxic cells in eliminating viral pathogens, this profound loss in lymphocytes may impair patients' ability to eliminate the virus. IL-7 is a pleiotropic cytokine that is obligatory for lymphocyte survival and optimal function.

Methods: We conducted a prospective, double-blind, randomized, placebo-controlled trial of CYT107, recombinant human IL-7, in 109 critically ill, patients with lymphopenia who have COVID-19. The primary endpoint was to assess CYT107's effect on lymphocyte recovery with secondary clinical endpoints including safety, ICU and hospital length-of-stay, incidence of secondary infections, and mortality.

Results: CYT107 was well tolerated without precipitating a cytokine storm or worsening pulmonary function. Absolute lymphocyte counts increased in both groups without a significant difference between CYT107 and placebo. Patients with COVID-19 receiving CYT107 but not concomitant antiviral medications, known inducers of lymphopenia, had a final lymphocyte count that was 43% greater than placebo (P = 0.067). There were significantly fewer treatment-emergent adverse events in CYT107 versus placebo-treated patients (P < 0.001), consistent with a beneficial drug effect. Importantly, CYT107-treated patients had 44% fewer hospital-acquired infections versus placebo-treated patients (P = 0.014).

Conclusion: Given that hospital-acquired infections are responsible for a large percentage of COVID-19 deaths, this effect of CYT107 to decrease nosocomial infections could substantially reduce late morbidity and mortality in this highly lethal disease. The strong safety profile of CYT107 and its excellent tolerability provide support for trials of CYT107 in other potential pandemic respiratory viral infections.

Trial registration: NCT04379076, NCT04426201, NCT04442178, NCT04407689, NCT04927169.

Funding: Funding for the trial was provided by RevImmune and the Cancer Research Institute.

Keywords: COVID-19; Clinical trials; Immunotherapy; Infectious disease.

Figure 1. Trial enrollment details.

Details on patient enrollment, allocation, stratification, and number analyzed are presented. Hospitalized patients with respiratory distress who were positive for SARS-CoV-2 were identified. Patients were further screened for eligibility as per inclusion and exclusion criteria.

Figure 2. CYT107 did not increase circulating levels of IL-6.

Blood samples were collected from patients at 6–7 hours and 20–22 hours after the first and second doses of CYT107 or placebo. (A and B) Results showed that circulating levels of IL-6 were not different in patients receiving CYT107 versus placebo and represent averaged data (A) and individually plotted data (B). See Supplemental Figures 3 and 4 for results for IL-10 and TNF-α. Data are shown as mean ± SEM. FR, France; US, United States, UK, United Kingdom; D0, day 0, D3, day 3.

Figure 3. Effects of CYT107 on ICU and hospital length of stay.

(A) Comparison of the number of ICU days in the CYT107 versus placebo group showed that there was no difference in the number of ICU days in the patients receiving CYT107 versus placebo — i.e., 31 days versus 45 days respectively (P = 0.3366) . Because of potential effects of antiviral agents to inhibit the putative beneficial effects of CYT107, the number of ICU days were also calculated for patients receiving CYT107 or but no antiviral drugs versus placebo treated patients who were not receiving antiviral drugs. (B) Patients receiving CYT107 but no antiviral medications did have a statistically significant decrease in their number of ICU days — i.e., 24 days versus 45 days respectively (P = 0.0482). (C) Results for hospital length of stay showed similar trends as ICU length of stay. Hospitalization days in patients receiving CYT107 versus placebo — i.e., 29 days versus 45 days, respectively (P = 0.4096). (D) Comparison of the days of hospitalization in CYT107-treated patients who were not receiving antiviral agents was close to statistical significance at 28 days versus 45 days for CYT107- and placebo-treated patients respectively (P = 0.0623). Survival, ICU, and hospital length of stay were analyzed with a log-rank (Mantel-Cox) test to compare the curves. Note that patients who died during their ICU or hospital stay were censored at day 45 even if they died prior to that day. Comparison of ICU and hospital length of stay was performed using the log-rank (Mantel-Cox) test. *P < 0.05.

Figure 4. No difference in mortality in CYT107- versus placebo-treated patients.

The mortality rate was calculated at day 30 or hospital discharge if patients left the hospital prior to day 30. (A) The overall mortality was 32.7% in CYT107-treated patients (n = 55) versus 38.9% in placebo-treated patients (n = 54) and was not statistically different (P = 0.59) (Table 3). (B) Because of potential effects of antiviral agents to inhibit the putative beneficial effects of CYT107, mortality rates were also determined separately for patients receiving CYT107 but no antiviral drugs and similarly for placebo-treated patients who were not receiving antiviral drugs. Mortality for CYT107-treated patients (n = 22) and placebo-treated patients (n = 23) was 22.7% and 47.8%, respectively (P = 0.17) (Table 3). Comparison of mortality was performed using the log-rank (Mantel-Cox) test.