Intravescical instillation of Calmette-Guérin bacillus and COVID-19 risk

ABSTRACT

It has been theorized that Calmette-Guérin bacillus may prevent or reduce the severity of COVID-19 through a nonspecific stimulation of the immune system. A preliminary assessment of SARS-CoV-2 infection rates and outcomes among 2803 individuals affected with high risk non-muscle-invasive bladder cancer and treated with intra-bladder instillation of BCG, showed no evidence of a protective effect. However, the interpretation of these data need some caution, due to the low prevalence of infection (<1%) observed within this population, along with the fact that intra-bladder administration cannot mirror the usual intradermal administration of BCG, in particular in patients partially immunocompromised. Confirmation by larger prospective studies is required.

KEYWORDS:

• BCG
• bladder Cancer
• Covid-19
• epidemiology

Commentary

Calmette-Guérin bacillus (BCG) may prevent or reduce the severity of COVID-19 infection through a nonspecific stimulation of the immune system by inducing an immune response against pathogens other than M. tuberculosis both at the level of innate and adaptive immunity.¹ Moreover, BCG administration prior to pandemic influenza vaccination (H1N1v) resulted in an accelerated and more pronounced antibody response.² Ecological studies have suggested that countries with mandatory childhood BCG vaccination have lower COVID-19 morbidity and mortality rates, notwithstanding comparisons were done at different points on the epidemic curve.³ WHO recommends BCG vaccine for COVID-19 only in randomized controlled trials, currently ongoing in high risk populations⁴

COVID-19 pandemic might particularly affect patients treated with intra-bladder BCG for high risk non-muscle-invasive bladder cancer (BCa), mostly elderly males with associated comorbidities favoring severe COVID-19 infection. However, since these patients might be protected by the BCG stimulus, analyses on COVID-19 incidence are warranted.⁵

We investigated SARS-CoV-2 infection rates and outcomes among BCG-treated BCa patients in the Veneto Region (Italy, 4.9 million inhabitants). Subjects alive on February 1, 2020 and treated with BCG instillations in the period 2015–2019 were identified by the corresponding Anatomical Therapeutic Chemical code. Infections ascertained by nasopharyngeal swabs, related hospitalizations, and outcomes were assessed up to May 14, 2020 by linkage with a centralized surveillance system including all COVID-19 cases. An adverse outcome was defined by intensive-care unit (ICU) admission, or death. Events in the BCG cohort were compared both with the overall regional population, and with a cohort of BCa patients hospitalized in 2015–2019 not treated with BCG. Standardized incidence ratios (SIR) with 95% confidence intervals were computed based on age and sex specific rates of the two reference populations.

The BCG cohort included 2,803 patients (84% males, 69% aged ≥70 years). Overall 19 cases of COVID-19 infection were registered, with 14 hospitalizations and 7 ICU admissions/deaths. The cohort of BCa patients not treated with BCG (N = 11,130) had similar demographic characteristics (and rates of COVID-19 related events (Table 1). Once comparing the BCG cohort with the overall regional population (rates shown in Table 2), the SIR for COVID-19 infection was 1.04 (95%CI 0.62–1.62), for hospital admissions 1.41 (0.77–2.36), and for adverse outcomes 1.32 (0.53–2.71). When patients with BCa without BCG treatment were taken as reference, the SIR was 0.96 (CI 0.57–1.49), 1.01 (0.55–1.70), and 0.95 (0.38–1.96) for infections, hospitalizations, and adverse outcomes, respectively.

Table 1. Patients treated with BCG in 2015–2019 and patients with a hospital diagnosis of bladder neoplasm without BCG treatment in Veneto, Italy: number of study subjects, COVID-19 infections, hospitalizations, and adverse outcomes

	Subjects in the cohort		COVID-19 infections		Hospitalizations		ICU admissions/ deaths
	BCG	No BCG	BCG	No BCG	BCG	No BCG	BCG	No BCG
Males
<60	200	960	1	7	-	1	-	-
60–69	507	1,754	3	5	1	4	-	1
70–79	980	3,304	10	22	8	18	4	9
80+	658	2,729	5	33	5	25	3	14
Total	2,345	8,747	19	67	14	48	7	24
Females
<60	45	307	-	2	-	-	-	-
60–69	110	515	-	1	-	1	-	-
70–79	168	717	-	1	-	1	-	1
80+	135	844	-	11	-	4	-	5
Total	458	2,383	-	15	-	6	-	6

Table 2. Rates of COVID-19 infection, hospitalization, and adverse outcomes in the overall population (n= 4,905,854) of the Veneto Region, Italy

		Rate x 1,000 pop
	Population, n	COVID-19 infections	Hospitalizations	ICU admissions/deaths
Males
<60	1,758,311	2.4	0.5	0.1
60–69	288,920	4.6	2.4	0.7
70–79	227,295	5.2	3.2	1.5
80+	125,257	11.2	6.8	4.4
Total	2,399,783	3.4	1.3	0.5
Females
<60	1,713,124	3.3	0.3	0.0
60–69	306,168	3.1	1.0	0.2
70–79	264,871	3.8	1.7	0.5
80+	221,908	14.4	4.9	3.1
Total	2,506,071	4.3	0.9	0.4

The present population-based data on 2,803 BCG-treated BCa patients showed no evidence of a protective effect. Risks of infection, hospitalization, and adverse outcomes overlapped with risks of untreated BCa patients and the regional population. Meanwhile, a report from Israel showed no difference in SARS-CoV-2 infection rates between birth cohorts (1979–1981 vs.1983–1985) submitted or not to neonatal BCG vaccination.⁶

In spite of a large number of recruited patients, this study had a limited power to detect a protective role of BCG treatment due to low overall rates of infection among BCa patients (<1%). Additionally, cancer patients ought to be considered as partially immunosuppressed individuals both at the mucosal and at the systemic level. Lastly, intra-bladder instillation of the bacillus – although repeated during the induction and maintenance courses – cannot mirror intradermal inoculation of an antigen such as BCG. The mucosal bladder environment is certainly different from that of derma, a district notoriously rich in dendritic cells, the most active antigen-presenting cells part of the innate immune defense involved in trained immunity and cross-talk with the adaptive immune response.

Therefore, some caution has to be adopted correlating BCG immune-reactivity with the anti- SARS-CoV-2 effect in our setting. The absence of a protective response associated to BCG treatment must be confirmed by larger prospective investigations.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.