Since December 2019, the coronavirus disease (COVID-19) has a major impact on the healthcare system and social life around the world. On 17 March 2020, a nationwide lockdown was implemented in France in order to prevent the disease from spreading (Cauchemez et al., 2020).

The COVID-19 pandemic and associated social restrictions have deteriorated the population’s psychological health, causing fears of infection, anxiety, and stress (French et al., 2020; Salari et al., 2020; Chen et al., 2021). It has been described as a traumatic stressor that may trigger symptoms of post-traumatic stress disorder (PTSD), especially in women and people at high risk of severe infection (Bridgland et al., 2021). Breast cancer (BC) patients might be particularly vulnerable to such psychological distress (Bargon et al., 2021). Owing to their weakened immune system and their poor survival outcomes when infected by the COVID-19 (Liang et al., 2020), cancer patients, especially female patients, have developed fears to be infected by the COVID-19 (Sigorski et al., 2020; Sutcuoglu et al., 2020). In addition, they have been deprived of social support from family and friends which led to social isolation and loneliness (Gallagher et al., 2020; Biagioli et al., 2021). Moreover, the pandemic induced unexpected and challenging problems to the routine delivery of BC treatment management and supportive healthcare (Chan et al., 2020; Soran et al., 2020). Numerous guidelines have been developed to adjust BC medical protocols (Curigliano et al., 2020; de Azambuja et al., 2020; Manoj Gowda et al., 2020). Surgical treatments were delayed when possible, different forms of treatment were modified, and monitoring visits were canceled, postponed, or turned into video calls. This led to fears of cancer care disruption (Guven et al., 2020).

The psychological health of BC patients has been described as very deteriorated in literature, as reported by a meta-analysis of Hashemi et al. (2020). Especially, PTSD symptoms are frequently observed at diagnosis and may persist over time for half of the patients (O’Connor et al., 2011; Voigt et al., 2017). Few studies have investigated the psychological health of BC during the COVID-19 pandemic, which may be considered as a potential additional stress factor. A study from the Netherlands observed deterioration in emotional functioning and moderate to severe loneliness in patients (Bargon et al., 2021). A British study suggested that the pandemic-induced disruption to scheduled oncology services had a negative impact on cognitive function (Swainston et al., 2020). A French study measured PTSD and anxiety in cancer patients in France following the first COVID-19-related lockdown and associated factors (Marino et al., 2022).

To our knowledge, only two studies explored PTSD symptoms in BC patients at the very beginning of the COVID-19 pandemic in China and found that BC experienced high rates of anxiety, depression, distress, and insomnia (Cui et al., 2020; Juanjuan et al., 2020). This study aimed to estimate the rate of PTSD symptoms among BC patients during the first COVID-19 lockdown in France and to identify the associated clinical and psychosocial factors.

Among the 304 patients with BC from the COVIPACT study, we included 253 patients who completed questionnaires on PTSD symptoms, QoL, insomnia, and cognition. Among them, 183 (72%) patients also completed the COVID-19 social and psychological experiences questionnaire.

In our study, we observed that 23% of BC patients had PTSD symptoms during the first lockdown in France. Fears relative to COVID-19 infection, feeling of isolation, and prescription or increased use of psychotropic drugs during lockdown (reported by 65, 48, and 27% of the patients, respectively) were associated with more frequent PTSD symptoms. Besides, we observed poor QoL, cognitive function, and sleep quality in patients with PTSD symptoms compared to patients without PTSD symptoms.

Several studies reported poor mental health and high level of anxiety and distress in the population during the COVID-19 pandemic. Two studies showed that 16.3 and 23.6% of the general population had PTSD symptoms during the COVID-19 pandemic (as measured with the IES-R Questionnaire) (Alkhamees et al., 2020; Tee et al., 2020). Generally, a systematic review reported that COVID-19 was associated with psychological distress, especially for women, with however a great heterogeneity in PTSD symptoms rates (from 7 to 53.8%) (Xiong et al., 2020). Among cancer patients, it was reported 36% of PTSD symptoms in lymphoma patients during the onset of COVID-19 in Italy (Romito et al., 2020) and 31% in an American study of mostly BC patients (Miaskowski et al., 2020). A French oncological survey identified moderate-to-severe PTSD in 14.7% of cancer patients during the first pandemic COVID-19 lockdown, with a higher rate of 18% in the breast or gynecological patients (Marino et al., 2022). Two Chinese studies that assessed the prevalence of PTSD symptoms very early in the course of the COVID-19 pandemic found 35.5 and 52.2% moderate to severe distress symptoms in BC patients (Cui et al., 2020; Juanjuan et al., 2020). Overall, our findings were comparable to those found in cancer patients’ studies that generally found a higher frequency of PTSD symptoms than that in the general population.

According to our study, half of BC patients felt isolated during the lockdown, two-third had fears relative to the COVID-19 infection, and both factors were associated with PTSD symptoms. In agreement, a qualitative study described that the lockdown was very stressful for BC patients because they had to limit their outings and interactions with family and friends, and because they were concerned about catching the virus, especially when going to the hospital (Savard et al., 2021). Social isolation, as measured by a dedicated scale on emotional and social loneliness, was also observed in 48% of BC patients in a prospective Dutch study (Bargon et al., 2021). In this study, they also reported a significant drop in the social functioning of actively treated patients during the COVID-19 when compared to a pre-COVID-19 period (Bargon et al., 2021). Regarding fears of COVID-19 infection, varying levels of worry were found in previous studies in cancer patients, ranging from 19.8 to 65.6% (Guven et al., 2020; Juanjuan et al., 2020). Especially, it has been shown in general European populations and cancer population that women had a significantly higher level of fear of COVID-19 than men (Reznik et al., 2020; Sigorski et al., 2020). As a result, among cancer patients, the highest level of fear and anxiety was observed in patients with BC (Sigorski et al., 2020). However, it was suggested that actively treated patients with cancer were more concerned about cancer than about COVID-19 infection (Sigorski et al., 2020; Patni et al., 2021), both fears being closely linked. Indeed, some studies reported that BC patients were concerned about the efficacy of delayed/adapted therapies (Juanjuan et al., 2020). In other studies, BC patients interrupted their treatment because of fears of contracting COVID-19 in hospitals, resulting in changes and delays in their treatments (Alipour et al., 2020; Papautsky and Hamlish, 2020; Villarreal-Garza et al., 2021).

A third of our patients had a prescription or an increased use of psychotropic drugs. Several studies, including some in the general population, evaluated the use of psychotropic drugs during the period of the pandemic. Two studies conducted at the pandemic onset (between February and March 2020) in the United Kingdom and the United States found a considerable increase in the use of psychotropic drug prescriptions compared to 2019 (Stettin, 2020; Rabeea et al., 2021). Moreover, when the lockdown was declared in the United Kingdom, medication interventions increased by two per week, with hypnotics and benzodiazepines being the most commonly prescribed psychotropic medications (Rauf et al., 2021). Moreover, García-Fernández et al. (2020) found that the use of anxiolytics was associated with higher emotional distress in the elderly, which is in line with our results.

In the entire COVIPACT cohort, cancer patients who experienced an adjustment in medical oncology practices were found to be more vulnerable to PTSD symptoms (Joly et al., 2021). In this study focused on the sub-population of BC patients, we found that 27% of patients experienced adjustment in medical oncology practices during the lockdown, but it was not associated with more PTSD symptoms. This discrepancy in our global population might relate to the type of adjustment that was implemented. In BC patients, the most common adjustment was the implementation of teleconsultation, which may be less troubling for patients than rescheduled or rearranged treatments. Indeed, two studies aiming to assess the relationship between delaying treatment and deterioration of cancer patients’ mental health observed that treatment delay or interruption was among the dominant factors contributing to anxiety and depression (Chen et al., 2020; Ciazynska et al., 2020).

Finally, our results indicated that patients with PTSD symptoms presented worse QoL, more cognitive complaints, and more insomnia compared to patients without PTSD symptoms. Similarly, Miaskowski et al. (2020) found that cancer patients with PTSD symptoms reported clinically meaningful levels of depressive symptoms, anxiety, pain, cognitive complaints, and sleep disturbance.

There were some limitations in this study. First, this is a cross-sectional analysis of the COVIPACT study in which no data were collected before lockdown, which precluded to properly assess that PTSD symptoms emerged from the COVID-19 (and associated social restrictions) stressor. O’Connor et al. (2011) showed that receiving a BC diagnosis could also be a significant traumatic experience and that many women experienced persistent cancer-related PTSD symptoms. This suggests that PTSD symptoms could be associated with cancer rather than with COVID-19. However, in our study, only factors related to COVID-19 (e.g., fears of infection and social isolation), but no cancer-related factors (e.g., stage of cancer; time since diagnosis), were associated with PTSD symptoms, which is in support of the contribution of the COVID-19 context to the deterioration of psychological health in cancer patients. Second, the IES-R Questionnaire is not a formal diagnostic tool, but it is widely used to detect symptoms of psychological distress. Finally, although most patients-reported outcomes were assessed using validated questionnaires, social isolation was assessed using a single question of a non-validated questionnaire, rather than a specific and validated scale (Sarason et al., 1987). Despite these limitations, our study is one of few studies that examined specifically PTSD symptoms in BC patients and explored their association with both clinical factors (collected from medical records) and psychosocial factors during the first COVID-19 lockdown. The recruitment period coincides specifically with the date of the first lockdown (from 17 March to 3 May 2020). In the beginning, only 7,730 cases of COVID-19 were observed in France but the spread of the virus and the high rate of death (175 among 7,730) may contribute to the severity of anxiety and PTSD. Moreover, the PTSD assessment has been performed around 1 month after the start of the lockdown, which is a sufficient time duration to observe the true effect of lockdown on PTSD symptoms.

Overall, according to our study, 23% of BC patients experienced PTSD symptoms during the first COVID-19 lockdown in France. While no clinical factors were associated with PTSD symptoms, patients who had fears relative to COVID-19, feeling of social isolation, and prescription or increased use of psychotropic drugs were more likely to express PTSD symptoms, supporting that the COVID-19 pandemic, beyond cancer, may contribute to psychological distress in this BC population. This study can contribute to suggest additional psychological support during this long period of pandemic. COVIPACT 2 (NCT04747249), a study including the cancer patients of COVIPACT study identified with a moderate to severe level of PTSD, is underway to assess the benefit of psychological interventions to reduce them.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by ref. 220 C07; South Mediterranean II Committee for the Protection of Persons. The patients/participants provided their written informed consent to participate in this study.

FY: bibliography, statistical analysis, and writing the manuscript. JL: conceptualization, methodology, results interpretation, help for redaction, supervision, and submission. OR: resources and project administration. AM and ML: resources. J-MG and AL: project administration. BC: conceptualization, methodology, funding acquisition, and supervision. FJ: conceptualization, results interpretation, help for redaction, and supervision. SL-A: bibliography, results interpretation, help for redaction, and supervision. All authors contributed to the article and approved the submitted version.