MATERIALS AND METHODS: A prospective cohort study of cancer patients and healthy individuals receiving vaccines was conducted in Malaysia. All participants were aged 18 or above at recruitment and received at least two doses of vaccine. We excluded patients who had missing serum antibody data post-first dose and post-second dose. Sociodemographic and clinical data were collected at baseline, prior to vaccination. Data on self-reported breakthrough infection was collected at six months. Multivariable linear mixed-effects regression models were used to investigate the association between the type of vaccine and serum IgG titer.

RESULTS: A total of 389 patients with solid (n=276, 71.0%) or hematologic cancers (n=113, 29.0%) were included, along with 246 healthy individuals. Most cancer patients received BNT162b2 (n=358, 92.0%), followed by AZ1222 (n=19, 4.9%) and Coronavac (n=12, 3.1%). Most healthy individuals received BNT162b2 (n=151, 61.4%), followed by Coronavac (n=95, 38.6%). Vaccination, after adjustment for confounders (pre-vaccine infection, age, ethnicity, comorbidity, timepoint, income, cancer type, and booster), with Coronavac was associated with lower log IgG titer (-3.09 U/ml, 95% confidence interval=-4.37 to -1.80, p<0.01) than that of BNT162b2 in patients with cancer and also lower log IgG titer (-2.64 U/ml, 95% confidence interval=-2.97 to -2.30, p<0.01) than that of BNT162b2 in healthy individuals. No effect modification by sex was observed. Among the cancer cohort, 76 patients (19.5%) reported breakthrough infections after vaccination, while 33 (13.4%) participants in the healthy cohort reported breakthrough infections after vaccination. Coronavac was associated with greater odds of breakthrough infection among healthy individuals (odds ratio=7.34 compared to BNT162b2, confidence interval=1.40 to 33.49, p=0.02).