Abstract
Background
The risks of hospital admission for COVID-19-related conditions and all-cause death of SARS-CoV-2 infected cancer patients were investigated according to vaccination status.
Methods
A population-based cohort study was carried out on 9754 infected cancer patients enrolled from January 1, 2021 to June 30, 2022. Subdistribution hazard ratio (SHRs) or hazard ratios (HRs) with 95 % confidence intervals (CI), adjusted for sex, age, comorbidity index, and time since cancer incidence, were computed to assess the risk of COVID-19 hospital admission or death of unvaccinated vs. patients with at least one dose of vaccine (i.e., vaccinated).
Results
2485 unvaccinated patients (25.5 %) were at a 2.57 elevated risk of hospital admission (95 % CI: 2.13–2.87) and at a 3.50 elevated risk of death (95 % CI: 3.19–3.85), as compared to vaccinated patients. Significantly elevated hospitalizations and death risks emerged for both sexes, across all age groups and time elapsed since cancer diagnosis. For unvaccinated patients, SHRs for hospitalization were particularly elevated in those with solid tumors (SHR = 2.69 vs. 1.66 in patients with hematologic tumors) while HRs for the risk of death were homogeneously distributed. As compared to boosted patients, SHRs for hospitalization and HRs for death increased with decreasing number of doses.
Conclusions
Study findings stress the importance of SARS-CoV-2 vaccines to reduce hospital admission and death risk in cancer patients.
Keywords: SARS-CoV-2, Vaccination, Cancer, Northeastern Italy, Hospital admission, Death
1. Introduction
It has been well documented since the early phase of the COVID-19 pandemic that cancer patients with SARS-CoV-2 infection were at increased risk of hospital admission and death, particularly elderly ones, those with major medical comorbidities and those under immunosuppressive anti-cancer therapies [1], [2], [3], [4], [5]. An elevated 30-day mortality was documented among 928 patients with COVID-19 and cancer enrolled in The COVID-19 and Cancer Consortium (CCC19) cohort [5], while among 1557 cancer patients with COVID-19 enrolled in the OnCovid multicenter study, 72.2 % of those with sequelae required hospitalization, as compared to 41.2 % of those without sequelae [1]. According to early evidence, cancer patients were among the prioritized population groups for vaccination against SARS-CoV-2 infection.
The impact of SARS-CoV-2 infection on clinical outcomes of cancer patients in northeastern Italy has been already investigated [2]. In this investigation, we updated the previous analysis by focusing on the impact of SARS-CoV-2 vaccination on infected cancer patients. The study objective was to assess the risk of COVID-19 hospitalization or death in infected cancer patients according to SARS-CoV-2 vaccination status.
2. Methods
2.1. Study design and population
A population-based, cohort study was conducted using data derived from healthcare databases of the Friuli Venezia Giulia region (1,210,000 inhabitants, north-eastern Italy). These health-related databases cover - as previously published [6] - the whole resident population and provide information on patients’ personal data (e.g., sex, birth date, place of residence, vital status) and medical care history (e.g., hospital discharges, outpatient care, histopathological reports, and medical fee waivers). To assess cancer history, cancer patients diagnosed before December 31, 2020 were identified from the regional population-based cancer registry, while those diagnosed from January 1, 2021 to June 30, 2022 were identified from both the regional databases of hospital discharge records and the pathology reports. De-identified data from these databases were linked through an encrypted code, which is changed every six months in order to guarantee anonymity. The study was approved by the Bioethics Committee of the Veneto Regional Authority (Protocol no. 245343/2020).
Eligible individuals were all residents of the Friuli Venezia Giulia region who met the following criteria: (a) a positive SARS-CoV-2 nasopharyngeal swab test analyzed by means of real-time reverse transcription-polymerase chain reaction (RT-PCR) between January 1, 2021 and June 30, 2022; (b) a history of cancer diagnosis other than non-melanoma skin cancers. Information on SARS-CoV-2 test results was identified from the laboratory database, which contains all the microbiological testing performed in regional facilities. For individuals with multiple tests, those with at least one positive result were considered as SARS-CoV-2 positive at the date of the first positive test. For cancer patients, the primary cancer site was retrieved together with the time elapsing between the cancer diagnosis and the date of positive testing for SARS-CoV-2 infection. For patients with multiple cancers, the most recent cancer diagnosis before SARS-CoV-2 testing was considered.
All eligible individuals were categorized as ‘vaccinated’ if they had received at least one dose of SARS-CoV-2 vaccine, before their positive testing, regardless of the type of vaccine administered among the three made available in the study period in the Friuli Venezia Giulia region, or ‘unvaccinated’ if they had not. The main outcome of the analysis was death and hospital admission to a COVID-19 ward. In accordance with literature data, we focused on cancer patients admitted to a COVID-19 ward because such admission represents a further specific health risk for cancer patients. ICU admission was not considered a study outcome, and was not an exclusion criterion.
2.2. Statistical analysis
The cohort of SARS-CoV-2-positive cancer patients was examined to compare the clinical outcomes of SARS-CoV-2 infection (i.e., admission in a COVID-19 ward or death) in unvaccinated individuals vs. vaccinated ones. Risks of COVID-19-related hospital admission and of death for any cause were evaluated overall and in strata of selected characteristics. Person-time at risk of COVID-19-related hospital admission was computed as the time elapsed from the date of positive SARS-CoV-2 test to the date of admission in a COVID-19 hospital, to death, or to October 31, 2022 (i.e., date of the last follow-up), whichever came first. To account for death as a competing event, Fine and Gray’s regression models [7] were used for estimating subdistribution hazard ratios (SHRs) of COVID-19 hospital admission, with corresponding 95 % confidence intervals (CIs). In these models, which allow estimating the effect of covariates on the cumulative incidence function for the event of interest, subjects who experience the competing event (death) remain in the risk set (rather than being censored).
Person-time at risk of death for any cause was computed from the date of positive SARS-CoV-2 test to the date of death, or to October 31, 2022, whichever came first. Hazard ratios (HRs) of death for all causes and corresponding 95 % CIs were estimated using Cox proportional hazards models [8].
All models were adjusted for sex, age, Elixhauser Comorbidity Index [9], and time since cancer diagnosis.
3. Results
The overall population of 202,621 individuals who tested positive for SARS-CoV-2 infection during the study period included 9754 (4.8 %) individuals with a history of cancer diagnosed before testing ( Fig. 1). Among these cancer patients, 7269 (74.5 %) received at least one dose of vaccine, while 2485 (25.5 %) were unvaccinated (Fig. 1). The median follow-up time from positive test to study closure was 250 days among vaccinated, and 262 days among the 2485 unvaccinated cancer patients.
Fig. 1.
Flow chart of the study population.
The analyzed clinical outcomes of SARS-CoV-2 infection among the 9754 cancer patients were hospitalization in a COVID-19 ward and all-cause death among unvaccinated individuals, as compared to vaccinated ones ( Table 1). During the period of observation, 22.8 % of unvaccinated individuals and 10.2 % of those vaccinated were hospitalized in a COVID-19 ward. Based on the Fine-Grey model, the unvaccinated status was associated with a 2.6-fold higher (SHR = 2.57, 95 % CI: 2.13–2.87) risk of hospitalization among subjects who were either event-free or who had experienced the competing event (death). This association was documented in women (SHR=2.86), in men (SHR = 2.41), in all age groups (SHRs ranged from 1.75 for those aged < 60 years to 3.06 for ages between 75 and 84 years), and in each time interval elapsed between cancer diagnosis and testing (SHRs ranges from 2.0 to 2.8). When considering cancer type (Table 1), all types of solid tumors diagnosed in unvaccinated patients were significantly associated with a higher risk of hospitalization, particularly uterus (SHR = 5.65), head and neck (SHR = 4.81), kidney (SHR = 3.55), and lung (SHR = 2.88) cancers. The risk of hospitalization in unvaccinated patients with hematological neoplasms was 1.7-fold higher overall (95 % CI: 1.14–2.42) than in vaccinated ones. However, a statistically significant increased risk of hospitalization was documented only among those with multiple myeloma (SHR = 2.68) (Table 1).
Table 1.
Subdistribution hazard ratios (SHR) for COVID-19 hospital admission and hazard ratios (HR) for death with corresponding 95 % confidence intervals (CI) among 9754 cancer patients infected with SARS-CoV-2 according to vaccination status, in strata of selected variables. Friuli Venezia Giulia, 2021–2022.
| Vaccinated |
Unvaccinated |
SHR of hospitalization (95 % CI)a,c | HR of death (95 % CI)b,c | |||||
|---|---|---|---|---|---|---|---|---|
| Total | COVID-19 hospitalization | Death | Total | COVID-19 hospitalization | Death | |||
| N | N (%) | N (%) | N | N (%) | N (%) | |||
| All | 7269 | 739 (10.2) | 1020 (14.0) | 2485 | 566 (22.8) | 869 (35.0) | 2.57 (2.13–2.87) | 3.50 (3.19–3.85) |
| Sex | ||||||||
| Women | 3828 | 251 (6.6) | 429 (11.2) | 1414 | 241 (17.0) | 373 (26.4) | 2.86 (2.40–3.41) | 3.16 (2.72–3.63) |
| Men | 3441 | 488 (14.2) | 591 (17.2) | 1071 | 325 (30.3) | 496 (46.3) | 2.41 (2.10–2.76) | 4.00 (3.54–4.51) |
| Age (years) at infection | ||||||||
| < 60 | 1952 | 56 (2.9) | 50 (2.6) | 784 | 38 (4.8) | 49 (6.3) | 1.75 (1.16–2.65) | 2.54 (1.71–3.78) |
| 60–74 | 2254 | 184 (8.2) | 221 (9.8) | 731 | 158 (21.6) | 206 (28.2) | 2.88 (2.34–3.56) | 3.46 (2.86–4.19) |
| 75–84 | 1958 | 288 (14.7) | 363 (18.5) | 587 | 228 (38.8) | 330 (56.2) | 3.06 (2.58–3.62) | 4.48 (3.82–5.17) |
| ≥ 85 | 1105 | 211 (19.1) | 386 (34.9) | 383 | 142 (37.1) | 284 (74.2) | 2.09 (1.70–2.58) | 3.24 (2.80–3.82) |
| Time from cancer diagnosis (years) | ||||||||
| < 1 | 868 | 115 (13.2) | 189 (21.8) | 320 | 94 (29.4) | 182 (56.9) | 2.34 (1.79–3.06) | 3.87 (3.14–4.76) |
| 1–2 | 638 | 69 (10.8) | 103 (16.1) | 239 | 57 (23.8) | 111 (46.4) | 2.30 (1.63–3.26) | 3.67 (2.80–4.81) |
| 3–5 | 576 | 58 (10.1) | 80 (13.9) | 186 | 32 (17.2) | 67 (36.0) | 1.96 (1.29–2.97) | 4.15 (2.97–5.77) |
| ≥ 6 | 5187 | 495 (9.5) | 684 (13.2) | 1740 | 383 (22.0) | 509 (29.3) | 2.77 (2.40–3.13) | 3.43 (3.06–3.87) |
| Cancer type/site | ||||||||
| Solid tumors | 6776 | 658 (9.7) | 943 (13.9) | 2312 | 525 (22.7) | 809 (35.0) | 2.69 (2.40–3.02) | 3.65 (3.32–4.01) |
| Breast | 1723 | 91 (5.3) | 152 (8.8) | 570 | 77 (13.5) | 111 (19.5) | 2.79 (2.06–3.78) | 2.95 (2.30–3.78) |
| Prostate | 1089 | 142 (13.0) | 152 (14.0) | 291 | 99 (34.0) | 142 (48.8) | 2.76 (2.14–3.57) | 4.49 (3.56–5.67) |
| Colon-rectum | 782 | 104 (13.3) | 152 (19.4) | 282 | 86 (30.5) | 115 (40.8) | 2.66 (2.01–3.53) | 2.99 (2.33–3.82) |
| Melanoma, skin | 524 | 33 (6.3) | 32 (6.1) | 165 | 16 (9.7) | 22 (13.3) | 2.04 (1.16–3.59) | 3.77 (2.12–6.56) |
| Thyroid | 321 | 11 (3.4) | 5 (1.6) | 115 | 9 (7.8) | 11 (9.6) | 2.48 (1.02–6.04) | 5.21 (1.80–17.0) |
| Lung | 260 | 35 (13.5) | 91 (35.0) | 124 | 46 (37.1) | 100 (80.6) | 2.88 (1.84–4.51) | 3.76 (2.79–5.07) |
| Kidney | 286 | 31 (10.8) | 36 (12.6) | 94 | 27 (28.7) | 31 (33.0) | 3.55 (2.17–5.81) | 5.32 (3.16–8.95) |
| Bladder | 242 | 37 (15.3) | 49 (20.2) | 74 | 22 (29.7) | 48 (64.9) | 1.89 (1.11–3.23) | 3.62 (2.40–5.45) |
| Uterus | 205 | 9 (4.4) | 17 (8.3) | 83 | 17 (20.5) | 18 (21.7) | 5.65 (2.47–12.9) | 3.57 (1.81–7.05) |
| Head and neck | 194 | 19 (9.8) | 35 (18.0) | 54 | 21 (38.9) | 23 (42.6) | 4.81 (2.61–8.87) | 4.81 (2.72–8.34) |
| Others | 1150 | 146 (12.7) | 222 (19.3) | 460 | 105 (22.8) | 188 (40.9) | 2.13 (1.66–2.73) | 4.63 (3.59–6.03) |
| Hematological tumors | 493 | 81 (16.4) | 77 (15.6) | 173 | 41 (23.7) | 60 (34.7) | 1.66 (1.14–2.42) | 3.37 (2.37–4.76) |
| Leukemia | 175 | 30 (17.1) | 30 (17.1) | 73 | 15 (20.5) | 28 (38.4) | 1.38 (0.75–2.56) | 3.36 (1.98–5.69) |
| Non-Hodgkin lymphoma | 112 | 19 (17.0) | 21 (18.8) | 49 | 15 (30.6) | 20 (40.8) | 1.61 (0.81–3.23) | 2.48 (1.28–4.77) |
| Hodgkin lymphoma | 105 | 6 (5.7) | 3 (2.9) | 34 | 3 (8.8) | 2 (5.9) | 2.00 (0.52–7.66) | – |
| Multiple myeloma | 101 | 26 (25.7) | 23 (22.8) | 17 | 8 (47.1) | 10 (58.8) | 2.68 (1.21–5.92) | 4.47 (1.90–9.80) |
Estimated using Fine-Gray models adjusted for sex, age, Elixhauser comorbidity index, time from cancer incidence.
Reference category: vaccinated cancer patients.
Estimated using Cox proportional hazard models adjusted for sex, age, Elixhauser comorbidity index, time from cancer incidence.
The proportion of deceased patients was 14.0 % among the 7269 vaccinated ones and 35.0% among the 2485 unvaccinated individuals (HR = 3.50, 95 % CI: 3.19–3.85) (Table 1). The increased risk of death emerged in all the considered subgroups of patients. Particularly elevated statistically significant increased risks (HRs ≥ 4) were noted for unvaccinated men (HR = 4.00), for those aged 75–84 years (HR = 4.48), for those with cancers of prostate (HR = 4.49), thyroid (HR = 5.21), kidney (HR = 5.32), head and neck (4.81), multiple myeloma (HR = 4.47), and for unvaccinated individuals with cancer diagnosed from 3 to 5 years before testing (HR = 4.15).
Table 2, Table 3 show the estimated SHRs of hospitalization in a COVID-19 ward (Table 2) and HRs of death from any cause (Table 3), according to the number of received doses. As compared with study subjects who received three doses (n = 3889 -i.e., boosted subjects), the risk of hospitalization in a COVID-19 ward was 1.72-fold higher in those who received two doses (n = 2875) (95 % CI: 1.48–2.01), 2.52-fold higher in those who received one dose (n = 505) (95 % CI: 1.98–3.21), and 3.48-fold higher in unvaccinated ones (95 % CI: 3.04–3.98) (Table 2). Such pattern of risk appeared more marked in patients with solid tumors (SHRs ranging from 1.9 -two doses vs. 3 doses- to 3.8- unvaccinated vs. 3 doses) than in those with hematological tumors. Similarly, the risk of all-cause death decreased with increasing number of doses received (Table 3): compared to those vaccinated with three doses, the HRs were 2.65 (95 % CI: 2.32–3.04) for cancer patients vaccinated with two doses, 3.62 (95 % CI: 2.91–4.46) for one dose, and 6.03 (95 % CI: 5.33–6.83) for zero doses. This pattern of risk was consistent across all the considered strata.
Table 2.
Subdistribution hazard ratios (SHR) for COVID-19 hospital admission with corresponding 95 % confidence intervals (CI) among 9754 cancer patients infected with SARS-CoV-2 according to vaccine doses, in strata of selected variables. Friuli Venezia Giulia, 2021–2022.
| Number of vaccine doses, SHR (95 % CI)a |
||||
|---|---|---|---|---|
| 3b | 2 | 1 | 0 | |
| Characteristics | (N = 3889) | (N = 2875) | (N = 505) | (N = 2485) |
| All | 1 | 1.72 (1.48–2.01) | 2.52 (1.98–3.21) | 3.48 (3.04–3.98) |
| Sex | ||||
| Men | 1 | 1.64 (1.25–2.15) | 2.98 (2.05–4.34) | 3.90 (3.10–4.92) |
| Women | 1 | 1.78 (1.48–2.14) | 2.25 (1.64–3.08) | 3.23 (2.73–3.83) |
| Age (years) at infection | ||||
| < 60 | 1 | 1.23 (0.68–2.21) | 2.69 (1.28–5.61) | 2.22 (1.29–3.81) |
| 60–74 | 1 | 1.60 (1.17–2.19) | 2.15 (1.31–3.51) | 3.90 (2.93–5.21) |
| 75–84 | 1 | 1.94 (1.52–2.47) | 2.85 (1.92–4.24) | 4.28 (3.45–5.29) |
| ≥ 85 | 1 | 1.67 (1.26–2.20) | 2.29 (1.46–3.60) | 2.66 (2.08–3.40) |
| Time from cancer diagnosis (years) | ||||
| < 1 | 1 | 1.21 (0.83–1.77) | 1.07 (0.52–2.21) | 2.54 (1.84–3.52) |
| 1–2 | 1 | 1.78 (1.01–2.89) | 1.21 (0.43–3.39) | 3.00 (1.94–4.64) |
| 3–5 | 1 | 2.19 (1.27–3.78) | 3.09 (1.29–7.43) | 2.92 (1.73–4.93) |
| ≥ 6 | 1 | 1.80 (1.49–2.18) | 3.11 (2.35–4.12) | 3.85 (3.26–4.56) |
| Cancer type/sitec | ||||
| Solid tumors | 1 | 1.86 (1.58–2.18) | 2.64 (2.05–3.40) | 3.79 (3.28–4.38) |
| Breast | 1 | 1.91 (1.22–3.00) | 3.35 (1.18–6.20) | 4.13 (2.77–6.14) |
| Prostate | 1 | 2.23 (1.59–3.13) | 1.85 (0.93–3.67) | 4.11 (2.96–5.70) |
| Colon-rectum | 1 | 1.60 (1.04–2.47) | 5.18 (3.12–8.59) | 3.86 (2.65–5.60) |
| Melanoma, skin | 1 | 2.36 (1.13–4.92) | 3.38 (1.08–10.50) | 3.05 (1.55–6.01) |
| Lung | 1 | 2.28 (1.08–4.84) | 5.91 (1.86–18.70) | 5.64 (2.95–10.80) |
| Kidney | 1 | 1.58 (0.85–3.10) | 1.28 (0.29–5.50) | 3.69 (1.99–6.86) |
| Bladder | 1 | 3.34 (1.58–7.04) | 2.15 (0.73–6.29) | 3.45 (1.67–7.11) |
| Uterus | 1 | 1.75 (0.40–7.13) | 3.17 (0.32–31.60) | 8.24 (2.46–27.50) |
| Head and neck | 1 | 1.62 (0.63–4.15) | 1.36 (0.28–6.56) | 6.13 (2.75–13.70) |
| Hematological tumors | 1 | 1.08 (0.67–1.73) | 2.33 (1.06–5.11) | 1.81 (1.19–2.74) |
| Leukemia | 1 | 1.00 (0.47–2.14) | 1.86 (0.42–8.26) | 1.43 (0.72–2.83 |
| Non-Hodgkin lymphoma | 1 | 1.12 (0.45–2.76) | 2.94 (1.45–5.93) | 1.72 (0.81–3.69) |
| Multiple myeloma | 1 | 1.04 (0.41–2.78) | 1.88 (0.46–7.73) | 2.92 (1.24–6.91) |
Estimated using Fine-Gray models adjusted for sex, age, Elixhauser comorbidity index, time from cancer incidence.
Reference category.
Cancers for which the total number of observed events was ≤ 20 are not reported in the table.
Table 3.
Hazard ratios (HR) for all-cause death with corresponding 95 % confidence intervals (CI) among 9754 cancer patients infected with SARS-CoV-2 according to vaccine doses, in strata of selected variables. Friuli Venezia Giulia, 2021–2022.
| Number of vaccine doses, HR (95 % CI)a |
||||
|---|---|---|---|---|
| 3b | 2 | 1 | 0 | |
| Characteristics | (N = 3889) | (N = 2875) | (N = 505) | (N = 2485) |
| All | 1 | 2.65 (2.32–3.04) | 3.62 (2.91–4.46) | 6.03 (5.33–6.83) |
| Sex | ||||
| Men | 1 | 2.98 (2.41–3.70) | 4.37 (3.18–5.93) | 6.06 (4.98–7.43) |
| Women | 1 | 2.52 (2.12–3.00) | 3.31 (2.43–4.42) | 6.52 (5.57–7.66) |
| Age (years) at infection | ||||
| < 60 | 1 | 1.47 (0.81–2.76) | 1.20 (0.39–3.06) | 3.17 (1.87–5.68) |
| 60–74 | 1 | 2.05 (1.53–2.79) | 2.25 (1.38–3.55) | 5.49 (4.17–7.32) |
| 75–84 | 1 | 2.71 (2.17–3.40) | 4.45 (3.08–6.28) | 7.66 (6.27–9.41) |
| ≥ 85 | 1 | 3.16 (2.55–3.93) | 4.98 (3.51–6.92) | 5.93 (4.85–7.28) |
| Time from cancer diagnosis (years) | ||||
| < 1 | 1 | 2.96 (2.15–4.12) | 4.15 (2.51–6.63) | 7.42 (5.54–10.10) |
| 1–2 | 1 | 2.49 (1.64–3.83) | 2.49 (1.12–4.99) | 6.08 (4.19–9.03) |
| 3–5 | 1 | 2.83 (1.76–4.60) | 4.50 (1.99–9.27) | 7.28 (4.73–11.40) |
| ≥ 6 | 1 | 2.66 (2.25–3.15) | 3.77 (2.87–4.90) | 5.87 (5.03–6.87) |
| Cancer type/sitec | ||||
| Solid tumors | 1 | 2.79 (2.43–3.22) | 4.14 (3.32–5.14) | 6.56 (5.76–7.48) |
| Breast | 1 | 3.63 (2.52–5.29) | 5.19 (3.06–8.55) | 6.33 (4.48–9.09) |
| Prostate | 1 | 2.59 (1.83–3.68) | 5.63 (3.25–9.33) | 7.78 (5.72–10.70) |
| Colon-rectum | 1 | 2.70 (1.90–3.86) | 5.20 (2.93–8.83) | 5.40 (3.88–7.60) |
| Melanoma, skin | 1 | 3.10 (1.49–6.59) | 3.10 (0.47–11.60) | 6.45 (3.22–13.40) |
| Lung | 1 | 3.70 (1.84–7.69) | 7.92 (1.18–31.50) | 10.90 (5.58–22.40) |
| Kidney | 1 | 3.07 (1.90–5.11) | 3.38 (1.47–7.15) | 7.76 (4.95–12.70) |
| Bladder | 1 | 4.50 (2.39–8.86) | 3.66 (1.15–9.94) | 7.88 (4.44–14.90) |
| Uterus | 1 | 1.47 (0.54–4.38) | 1.73 (0.09–10.30) | 4.55 (1.88–12.60) |
| Head and neck | 1 | 2.84 (1.35–6.26) | 2.72 (0.88–7.72) | 8.95 (4.30–19.70) |
| Hematological tumors | 1 | 2.18 (1.36–3.47) | 0.75 (0.18–2.12) | 4.38 (2.91–6.66) |
| Leukemia | 1 | 2.27 (1.10–4.76) | – | 4.56 (2.43–8.98) |
| Non-Hodgkin lymphoma | 1 | 1.23 (0.45–3.06) | – | 2.62 (1.29–5.45) |
| Multiple myeloma | 1 | 2.46 (0.98–6.10) | 0.86 (0.12–3.67) | 6.16 (2.37–15.6) |
Estimated using Cox proportional hazard models adjusted for sex, age, Elixhauser comorbidity index, time from cancer incidence.
Reference category.
Cancers for which the total number of observed events was ≤ 20 are not reported in the table.
4. Discussion
It is well documented that SARS-CoV-2 infection and related COVID-19 conditions will negatively impact the clinical outcomes of cancer patients, including hospital admission and death, as compared to the corresponding uninfected patients. Since their availability, the immunogenicity of SARS-CoV-2 vaccines has been demonstrated in the general population and in patients with several cancer sites/types [10], [11], [12]. Such efficacy includes the recent vaccines against the Omicron variant [13], an observation that has a specific significance in cancer patients, given their potential weakened vaccine-induced immunity [14]. Other studies, however, reported less effective immune responses after SARS-CoV-2 vaccination in cancer patients [15].
The results of this population-based analysis appear to indicate that SARS-CoV-2 infected cancer patients, a population group at particularly elevated risk of adverse events, take advantage from vaccination and booster doses to reduce the frequency of all-cause death and of hospitalization due to COVID-19-related conditions.
Approximately 25 % of the evaluated 9754 cancer patients were unvaccinated (according to the study definition) at the time of cohort enrollment, which is an inconsistent percentage compared to those reported by a subgroup analysis from the OnCovid multicenter study, where 72.0 % of cancer patients were unvaccinated in the Alpha-delta phase (Feb 27 2020–Dec 14 2021) and 12.4 % in the Omicron phase (Dec 15 2021–Jan 31 2022) [16]. A cross-sectional survey conducted in France to assess the acceptability of SARS-CoV-2 vaccination among cancer patients indicated that 16.6 % of the 999 study subjects would not accept vaccination [17]. Differences between our findings and those mentioned in the literature were also seen in the frequency of boosted cancer patients. Among the vaccinated cancer patients enrolled in the OnCovid multicenter study, 8.2 % of those in the Alpha-delta phase, and 55.6 % of cancer patients in the Omicron phase were boosted, as compared to a frequency of 53.5 % among the vaccinated cancer patients that emerged in our analysis. Although noteworthy differences in study design should be considered, it seems that our population-based results are in line with those documented in the Omicron phase of the COVID-19 pandemic by the 37 European oncology centers participating in the OnCovid study.
Study findings also documented that unvaccinated patients were at increased risk of hospitalization for COVID-19 conditions than vaccinated ones. In our analysis, the excess risk of hospital admission turned out to be of similar magnitude according to sex, it was more elevated in those aged 60 years or older and noticeable discrepancies emerged with regard to cancer type/site. Among patients with solid tumors, a particularly elevated risk of hospitalization was observed in unvaccinated patients with lung cancer -i.e., they were 5.6-fold more likely to be hospitalized than boosted lung cancer patients. This finding is in line with previous reports of elevated adverse events in infected cancer patients with respiratory tumors [18], and it is likely attributable to the already worsened functioning of the respiratory system. Even if the national guidelines considered cancer patients a vaccination priority, it is possible that some frail patients received different recommendations. The use of the severity index in the model adjustment could limit the impact of this on the SHRs. We also found that unvaccinated cancer patients with hematologic tumors were less likely to be protected from hospital admission than those with solid tumors. Hence, it seemed that vaccination against SARS-CoV-2 was less efficient in reducing the risk of hospital admission among patients with hematological tumors. This observation was particularly evident among patients with leukemia, among whom the vaccination did not confer a statistically significant advantage against hospital admission. The present findings appear to be in line with other reports focused on hematological tumors [2], [10], [19] and with reports highlighting that patients with hematological malignancies undergoing anti-CD20 treatments elicit a diminished immune response to vaccines [14], [20], [21]. With regard to the increased risk of death of unvaccinated cancer patients, it is worth stressing here that the overall excess risk was in accordance with the increased 28-day case-fatality rate reported for unvaccinated cancer patients in the OnCovid study (i.e., an HR of 3.84) [16].
Our analysis has strengths and limitations. Given the population-based nature of this registry study, selection biases were excluded and a large number of cancer patients could be included. On the other hand, several clinical information directly related to the risk of hospital admission of the study subjects were not available, most of all clinical stage and type of anti-cancer therapy. However, all estimates were adjusted for the Elixhauser comorbidity index in order to overcome, at least partially, such limitation. Furthermore, we computed SHRs/HRs according to the time elapsed since cancer diagnosis in order to minimize the heterogeneity among recent and long lasting cancer diagnoses. Another limitation is that we could not obtain information on the type of vaccine received, which may have introduced misclassification of the exposure. Nevertheless, to evaluate the impact of the definition of ‘vaccinated’ used for this analysis, we performed a sensitivity analysis considering as vaccinated all subjects who had received at least two doses of vaccine, which showed similar results (Supplementary Table 1). Finally, the present study does not include any comparison of the clinical outcomes of cancer vs. non-cancer individuals infected with SARS-CoV-2.
In conclusion, this study provides novel evidence supporting the role of SARS-CoV-2 vaccines in reducing the risk of COVID-19-related hospital admission and death in a large real-world population of patients with cancer. For these patients, in particular the more vulnerable ones with respiratory and hematological malignancies, SARS-CoV-2 vaccination and booster doses should be included among the key strategic tools against COVID-19.
CRediT authorship contribution statement
All authors substantially contributed to the concept and design of the present study and provided original data. The statistical analyses were carried out by Michele Gobbato and Martina Taborelli; the article was drafted by Michele Gobbato, Diego Serraino, and Martina Taborelli; Elena Clagnan and Stefania Del Zotto were responsible for the management of databases. All authors critically revised the article content and approved the final version submitted for publication.
Funding
This work was supported by Ricerca Corrente Line 2 and Ministry of Health (5 × 1000 2017 to CRO, Aviano). The funding source had no involvement in the study design, in the collection, analysis and interpretation of data, in the writing of the report, and in the decision to submit the article for publication.
Declaration of interest
None.
Acknowledgement
The authors thank Mrs Luigina Mei for editorial assistance.
Footnotes
Supplementary data associated with this article can be found in the online version at doi:10.1016/j.canep.2022.102318.
Appendix A. Supplementary material
Supplementary material.
References
- 1.Pinato D.J., Tabernero J., Bower M., et al. Prevalence and impact of COVId-19 sequelae on treatment and survival of patients with cancer who recovered from SARS-CoV-2 infection: evidence from the OnCovid retrospective, multicentre registry study. Lancet Oncol. 2021;22:1668–1680. doi: 10.1016/S1470-2045(21)00573-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Serraino D., Zucchetto A., Dal Maso L., et al. Prevalence, determinants, and utcones of SARS-CoV-2 infection among cancer patients. A population-based study in northern Italy. Cancer Med. 2021;10:7781–7792. doi: 10.1002/cam4.4271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Pinato D.J., Zambelli A., Aguilar-Company J., et al. Clinical portrait of the SARS-CoV-2 epidemic in European cancer patients. Cancer Discov. 2020;10:1465–1474. doi: 10.1158/2159-8290.CD-20-0773. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Grivas P., Khaki A.R., Wise-Draper T.M., et al. Association of clinical factors and recent anticancer therapy with COVID-19 severity among patients with cancer: a report from the COVID-19 and Cancer Consortium. Ann. Oncol. 2021;32(6):787–800. doi: 10.1016/j.annonc.2021.02.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kuderer N.M., Choueiri T.K., Shah D.P., et al. COVID-19 and Cancer Consortium. Clinical impact of COVID-19 on patients with cancer (CCC19): a cohort study. Lancet. 2020;395(10241):1907–1918. doi: 10.1016/S0140-6736(20)31187-9. Epub 2020 May 28. Erratum in: Lancet. 2020 Sep 12;396(10253):758. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Clagnan E., Gobbato M., Burba I., et al. COVID-19 infections in the Friuli Venezia Giulia Region (Northern Italy): a population-based retrospective analysis. Epidemiol. Prev. 2020;44(5–6 Suppl. 2):S323–S329. doi: 10.19191/EP20.5-6.S2.133. [DOI] [PubMed] [Google Scholar]
- 7.Fine J.P., Gray R.J. A Proportional hazards model for the subdistribution of a competing risk. J. Am. Stat. Assoc. 1999;94(446):496–509. [Google Scholar]
- 8.Cox D. Regression models and life tables (with discussion) J. R. Stat. Soc. Ser. B. 1972;34:187–220. [Google Scholar]
- 9.Elixhauser A., Steiner C., Harris D.R., et al. Comorbidity measures for use with administrative data. Med. Care. 1998:8–27. doi: 10.1097/00005650-199801000-00004. [DOI] [PubMed] [Google Scholar]
- 10.Oosting S.F., van der Veldt A.A.M., GeurtsvanKessel C.H., et al. mRNA-1273 COVID-19 vaccination in patients receiving chemotherapy, immunotherapy, or chemoimmunotherapy for solid tumours: a prospective, multicentre, non-inferiority trial. Lancet Oncol. 2021;22:1681–1691. doi: 10.1016/S1470-2045(21)00574-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Shepherd S.T.C., Fendler A., Au L., et al. Adaptive immunity to SARS-CoV-2 infection and vaccination in cancer patients: the CAPTURE study. Ann. Oncol. 2021;32(suppl_5):S1129–S1163. doi: 10.1016/annonc/annonc713. [DOI] [Google Scholar]
- 12.Nelli F., Giannarelli D., Fabbri, et al. Immunogenicity and early clinical outcome after two or three doses of SARS-CoV-2 mRNA-BNT162b2 vaccine in actively treated cancer patients: results from the prospective observational Vax-On-Third study. Ann. Oncol. 2022;33(7):740–742. doi: 10.1016/j.annonc.2022.04.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Shapiro L.C., Thakkar A., Campbell S.T., et al. Efficacy of booster doses in augmenting waning immune responses to COVID-19 vaccine in patients with cancer. Cancer Cell. 2022;40:3–5. doi: 10.1016/j.ccell.2021.11.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Addeo A., Shah P.K., Bordry N., et al. Immunogenicity of SARS-CoV-2 messenger RNA vaccines in patients with cancer. Cancer Cell. 2021;39:1091–1098. doi: 10.1016/j.ccell.2021.06.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Barrière J., Chamorey E., Adjtoutah Z., et al. Impaired immunogenicity of BNT162b2 anti-SARS-CoV-2 vaccine in patients treated for solid tumors. Ann. Oncol. 2021;32(8):1053–1055. doi: 10.1016/j.annonc.2021.04.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Pinato D.J., Aguilar-Company J., Ferrante D., et al. Outcomes of the SARS-CoV-2 omicron (B.1.1.529) variant outbreak among vaccinated and unvaccinated patients with cancer in Europe: results from the retrospective, multicentre, OnCovid registry study. Lancet Oncol. 2022;23:865–875. doi: 10.1016/S1470-2045(22)00273-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Barrière J., Gal J., Hoch B., et al. Acceptance of SARS-CoV-2 vaccination among French patients with cancer: a cross-sectional survey. Ann. Oncol. 2021;32(5):673–674. doi: 10.1016/j.annonc.2021.01.066. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Garassino M.C., Whisenant J.G., Huang L.C., et al. COVID-19 in patients with thoracic malignancies (TERAVOLT): first results of an international, registry-based, cohort study. Lancet Oncol. 2020;21:914–922. doi: 10.1016/S1470-2045(20)30314-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lee L.Y.W., Cazier J.B., Starkey T., et al. COVID-19 prevalence and mortality in patients with cancer and the effect of primary tumour subtype and patient demographics: a prospective cohort study. Lancet Oncol. 2020;21:1309–1316. doi: 10.1016/S1470-2045(20)30442-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Becerril-Gaitan A., Vaca-Cartagena B.F., Ferrigno A.S., et al. Immunogenicity and risk of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection after Coronavirus Disease 2019 (COVID-19) vaccination in patients with cancer: a systematic review and meta-analysis. Eur. J. Cancer. 2022;160:243e60. doi: 10.1016/j.ejca.2021.10.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Corti C., Antonarelli G., Scotté F., et al. Seroconversion rate after vaccination against COVID-19 in patients with cancer–a systematic review. Ann. Oncol. 2022;33(2):158–168. doi: 10.1016/j.annonc.2021.10.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary material.