The Multidimensional Assessment of COVID-19-Related Fears (MAC-RF) is an 8-item self-report measure, which is based on the theoretical premise that fear responses to COVID-19 involve different yet intertwined domains (i.e., bodily, relational, cognitive, and behavioural). In this multi-step study, we tested the psychometric properties of the French version of the MAC-RF and examined the reciprocal relationships among COVID-19-related fears. Data were collected in two French-speaking samples (N = 521 individuals from the community and N = 328 healthcare professionals). Internal reliability, convergent validity, construct validity, and internal structure of the MAC-RF were tested. The French version of the MAC-RF demonstrated good psychometric properties and a two-factor structure, with bodily and relational fears tapping into the first factor, and cognitive and behavioural fears tapping into the second factor. Healthcare professionals reported greater COVID-19-related fears than community participants. Correlation network analysis showed that fear for one’s own body and fear of taking action might increase the risk of experiencing other COVID-19-related fears. Limitations comprised the cross-sectional design of the study, risk of bias associated with self-report instruments, and use of online surveys. A careful assessment of different types of fear related to COVID-19 may have implications for prevention and clinical practice during the current coronavirus pandemic. The French version of the MAC-RF is valid and reliable and can thus be used for this purpose.