Basic characteristics of the study population
Eighty-two breast cancer patients between the ages of 31 and 76 years participated in the study. The study population consisted of 26 breast cancer patients undergoing adjuvant therapies, 26 breast cancer patients undergoing neoadjuvant therapies, and 30 patients undergoing palliative care. The distribution of current breast cancer stages according to the Union for International Cancer Control (UICC), and the subtype of breast cancer according to clinical surrogate definition, are shown in Table 2. The mean age of the female 82 breast cancer patients was 54.63 (±11.10) years. There were no significant differences in age between the different therapy statuses (p = 0.212); the mean age of patients undergoing adjuvant therapies was 58.00 (±11.73) years, neoadjuvant therapies 51.77 (±11.69) years, and palliative therapies 54.20 (±9.51) years. Other sociodemographic data are summarised in Table 3. In the analysis of comorbidities, 4/82 breast cancer patients (4.88%) suffered from a previous mental disorder. One patient was undergoing adjuvant therapy, and three were undergoing palliative care. We analysed medication as a proxy for other concomitant diseases, and there was an equal distribution between breast cancer patients undergoing adjuvant, neoadjuvant and palliative therapies.
According to the results from the first part of the COVID-19 pandemic questionnaire, 11 breast cancer patients (13.4%) reported an infection within the last 4 months. Ten suffered from respiratory tract infections and one from a urinary tract infection. 8/11 reported infections were diagnosed before the German lockdown on 16th March 2020. All throat swabs taken during routine clinical practice were negative for SARS-CoV-2. Levels of SARS-CoV-2 antibodies (immunoglobulin G, M, and A) were determined for all study patients. Only one patient had clinically relevant positive antibodies (IgG, IgA as well as IgM), but could not remember experiencing symptoms typical of a COVID-19 infection.
COVID-19 pandemic questionnaire
Table 1 represents the median values of the second part of the COVID-19 pandemic questionnaire for all breast cancer patients. The maximum median value was 3.00 (2.00–4.00), representing a high concern regarding the impact of the COVID-19 pandemic on life quality. Cronbach’s alpha was 0.88 for the “concern scale”, 0.86 for the “concern over time scale”, and 0.81 for the “impairment scale”. Questions 5–8 related to the timing of the pandemic and its associate restrictions. Figure 1 shows the significant results of the Friedman rank test (p = 0.001). Concerns significantly increased during three consecutive time points; time point “the first European patient got ill”, time point “the first European died” (p = 0.047), and time point “the number of SARS-CoV-2 infected persons raised” (p ≤ 0.0001). There was also a significant decrease in concern at the time point “German lockdown on March 16th 2020” (p = 0.019). All values of questions 5–8 significantly differ in the Wilcoxon test. There were no significant differences between adjuvant, neoadjuvant and palliative patients for sums of “concern scale” (10.00 [8.00–13.00] versus (vs.) 10.50 [8.50–14.00] vs. 10.00 [8.00–14.00]; p = 0.713), “concern over time scale” (8.00 [7.00–12.00] versus (vs.) 9.00 [8.00–12.00] vs. 9.00 [7.00–12.00]; p = 0.760), “impairment scale” (7.50 [5.00–10.00] versus (vs.) 8.00 [6.00–11.00] vs. 8.00 [7.00–10.00]; p = 0.977), and the median values of each question regarding COVID-19 worries. In addition, there were no significant differences between age (< 55 years and ≥ 55 years), civil status (single vs. married or long-term relationship), or children (no children vs. children vs. children already grown up) (data not shown).
Stress and coping inventory (SCI)
We used the stress and coping inventory to analyse stress, concern and coping strategies. For all participants, “stress caused by insecurity” was 19.00 (11.00–26.00), “stress caused by being overwhelmed” was 15.00 (10.00–20.00), and “stress caused by loss” was 11.00 (9.00–16.00). The mean value of “stress symptoms” was 23.00 (20.00–28.00). The value of coping strategies was 11.00 (10.00–12.00) for “positive coping”, 11.00 (9.00–12.00) for “active coping”, 13.00 (12.00–15.00) for “coping by support”, 10.00 (7.00–12.00) for “coping by believing in God or powers that be”, and 5.00 (4.00–7.00) for “coping by drinking alcohol and/or smoking”. Cronbach’s alphas were between 0.598 and 0.847 for 4/5 scales. The scale “coping by drinking alcohol and/or smoking” had a Cronbach’s alpha of 0.456. The inter-scale correlation is summarised in supplemental Table S1.
The SCI revealed that there was a significant difference in “positive coping” between different therapy regimens (p = 0.049; Kruskal-Wallis-Test). Breast cancer patients undergoing adjuvant therapy had significantly more positive coping strategies (12.00 [11.00–13.00]) than patients undergoing neoadjuvant therapy (11.00 [9.50–12.00]; p = 0.020). There were no other significant differences between the three therapy groups (Table 4).
According to age, patients younger than 55 years had significantly more “stress caused by insecurity” and “stress caused by being overwhelmed” than the older patients (23.00 [15.00–28.00] vs. 14.50 [9.00–21.50]; p = 0.019 and 18.00 [13.00–20.00] vs. 12.00 [9.00–16.50]; p = 0.014 respectively). “Stress caused by loss” and stress symptoms were slightly but not significantly higher for patients younger than 55 years (13.00 [10.00–18.00] vs. 10.50 [8.50–14.50], p = 0.061 and 26.00 [21.00–28.00] vs. 22.50 [18.00–28.00], p = 0.181 respectively). Regarding coping strategies, there were no significant differences between patients younger than 55 years and older patients (data not shown).
Single patients had significantly lower stress symptoms (21.00 [18.00–22.00]) than patients that were married or in a long-term relationship (25.00 [21.00–30.00], p = 0.004). However, “coping by support” was significantly higher in married patients or those in a long-term relationship (12.00 [10.00–14.00] vs. 14.00 [12.00–15.00], p = 0.039). No other scales of the SCI differed significantly in relation to civil status (data not shown).
Regarding children (no children vs. children vs. children already grown up), there were significant differences in “stress caused by insecurity” and “stress caused by being overwhelmed” between the stress groups (21.00 [12.00–30.00] vs. 25.00 [20.00–30.00] vs. 15.00 [9.00–21.00], p = 0.014 and 16.50 [11.00–22.00] vs. 20.00 [18.00–22.00] vs. 12.00 [9.00–16.00], p = 0.003 respectively). Patients with children reported significantly more stress than patients with grown up children (“stress caused by insecurity” 25.00 [20.00–30.00] vs. 15.00 [9.00–21.00], p = 0.003, “stress caused by being overwhelmed” 20.00 [18.00–22.00] vs. 12.00 [9.00–16.00], p = 0.001, “stress caused by loss” 15.00 [10.00–20.00] vs. 10.00 [8.00–16.00], p = 0.019). There were no significant differences in stress symptoms between coping strategies (data not shown).
Distress thermometer (DT)
The median of the NCCN® distress thermometer (DT) at initial breast cancer diagnosis before German lockdown was 5.00 (4.00–7.00) and did not differ significantly from the median of DT after German lockdown (5.00 [3.00–7.00], p = 0.260, Wilcoxon test). There were no significant differences in DT score for any subgroup (age, civil status, children, therapy status) between pre- and post-lockdown.
EORTC QLQ-C30 and QLQ-BR23
For all patients, median “global health status/quality of life (QL2)” was 58.33 (50.00–75.83)%, “physical functioning (PF2)” was 73.33 (60.00–93.33)%, “role functioning (RF)” was 66.67 (33.33–83.33)%, “emotional functioning (EF)” was 58.33 (41.67–75.00)%, “cognitive functioning (CF)” was 83.33 (50.00–100.00)%, “social functioning (SF)” was 66.67 (33.33–83.33)%, “fatigue (FA)” was 44.44 (22.22–66.67)%, “nausea and vomiting (NV)” was 0.00 (0.00–0.00)%, “pain (PA)” was 16.67 (0.00–50.00)%, “dyspnoea (DY)” was 0.00 (0.00–33.33)%, “insomnia (SL)” was 33.33 (0.00–66-67)%, “appetite loss (AP)” was 0.00 (0.00–33.33)%, “constipation (CO)” was 0.00 (0.00–33.33)%, “diarrhoea (DI)” was 0.00 (0.00–0.00)%, “financial difficulties (FI)” was 0.00 (0.00–33.33), “body image (BRBI)” was 66.67 (50.00–91.67)%, “sexual functioning (BRSEF)” was 16.67 (0.00–33.33)%, sexual enjoyment (BRSEE)” was 66.67 (33.33–100.00 ± 34.71)%, “future perspective (BRFU)” was 33.33 (0.00–66.67)%, “systemic therapy side effects (BRST)” was 28.57 (14.29–47.62)%, “breast symptoms (BRBS)” was 8.33 (0.00–25.00), “arm symptoms (BRAS)” was 22.22 (0.00–33.33) and “upset by hair loss (BRHL) was 66.67 (33.33–66.67). Cronbach’s alpha was between 0.666 and 0.902 with two exceptions (NV = 0.045 and BRST = 0.591). Supplemental Table S2 summarises the Spearman’ s rho test, showing the correlation between each item of EORTC QLQ-C30 and QLQ-BR23.
There were no significant differences in most of the items of EORTC QLQ-C30 and QLQ-BR23 with regards to children (no children vs. children vs. children, already grown up), civil status (single vs. married or long-term relationship) or age. However, physical functioning was significantly better (91.67 [66.67–100.00]%) for single patients than for those who were married or in a long-term relationship (73.33 [60.00–86.67]%, p = 0.023). Financial difficulties were significantly increased for patients younger than 55 years compared to the older patients (33.33 [0.00–66.67]% vs. 0.00 [0.00–33.33]%, p = 0.019).
Table 5 represents the comparison of patients undergoing different therapy types. There were significant differences in RF, SF, FA, PA and AL between patients undergoing adjuvant, neoadjuvant and palliative therapies. RF was highest for patients undergoing adjuvant therapy, in comparison to neoadjuvant and palliative therapies (83.33 [66.67–100.00]% vs. 66.67 [33.33–100.00]% vs. 66.67 [33.33–66.67]%, p = 0.035). The same pattern was observed for SF (66.67 [50.00–100.00]% vs. 66.67 [50.00–66.67]% vs. 33.33 [16.67–66.67]%, p = 0.044). Palliative patients had the most complaints regarding FA (27.78 [11.11–55.56]% vs. 33.33 [33.33–55.56] vs. 55.56 [33.33–77.78], p = 0.023), PA (16.67 [0.00–33.33] vs. 8.33 [0.00–33.33]% vs. 33.33 [16.67–66.67]%, p = 0.006) and AL (0.00 [0.00–0.00] vs. 0.00 [0.00–0.00] vs. 0.00 [0.00–66.67], p = 0.021). Mann-Whitney testing further revealed significant differences between adjuvant and palliative patients (RF p = 0.011; SF p = 0.012, FA p = 0.009, PA p = 0.015, AL p = 0.014), and neoadjuvant and palliative patients (PA p = 0.004; AL p = respectively 0.042).
Groups by concern
We did not observe any significant differences in pandemic-related stress levels between patients undergoing different therapy regimes. We therefore stratified patients based on their concerns regarding the COVID-19 pandemic. The minimum value of an answer on the COVID-19 pandemic scale was 1, and the maximum was 5. Questions 1–4 mainly represented concerns about risk of infection with COVID-19. The range of the sum of these four answers was 4 to 20 and allowed stratification into three groups; 27 patients (32.9%) with no concerns/only thoughts (sum range 4–8), 30 patients (36.6%) with a little concern (sum range 9–12), and 23 patients (28.0%) with concerns often/all the time.
This stratification was replicated using the SCI, revealing significant differences between these groups regarding stress caused by insecurity, being overwhelmed, loss, and stress symptoms (Table 6). The current DT further supported these significant differences. Patients with no concerns/only thoughts score had a DT value of 3.00 (2.00–5.00), patients with a little concern a value of 5.50 (3.50–7.00), and patients with concerns often/all the time scored 6.00 (5.00–7.00) (p ≤ 0.0001). Mann-Whitney U testing for group comparisons also revealed significant differences; no concerns/only thoughts versus a little concern had a p-value of 0.009, and no concerns/only thoughts versus concerns often/all the time had a p-value of ≤0.0001.
Table 7 shows the results of these three stratified subgroups for the EORTC QLQ-C30 and QLQ-BR23 questionnaires. The data revealed that the more COVID-19-related concern that was expressed, the lower the QL2, PF, RF, EF, CF, SF, BRBI and BRFU scores. These correlations were all statistically significant, with the exception of PF. Higher concerns were additionally associated with increased scores for FA, PA, DY, SL, CO, FI, BRST and BRAS. However, only, the associations with increased FA, SL and BRST were significant.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.