Immune checkpoint inhibition in the era of COVID-19.

The worldwide coronavirus pandemic continues to result in significant morbidity and mortality, with almost 24 million confirmed cases to date. Approximately 80% of patients have mild disease and do not require hospitalization. A key challenge facing the medical community is predicting which patients are at risk of developing severe disease, in order to initiate early supportive treatment and to facilitate enrolment into much needed prospective clinical trials, both crucial for developing and optimizing effective treatment strategies.

Patients with cancer have already been identified as having an increased risk of developing not only COVID‐19 infection, but also severe disease, both of which are associated with poorer clinical outcomes. Reassuringly, the increase in mortality from COVID‐19 infection in patients with cancer may be primarily related to age, sex and comorbidities rather than to the cancer itself. Furthermore, there was no increased mortality in patients receiving and those not receiving anticancer therapy.

Nevertheless, it is at least conceivable that the type of anticancer therapy may influence the risk and course of COVID‐19 infection in patients with cancer. Given the increasing use of immune checkpoint inhibition in Dermatology (metastatic melanoma, Merkel cell carcinoma and squamous cell carcinoma) we reviewed the current literature to determine the extent to which immune checkpoint inhibition has been associated with COVID‐19 infection.

We performed PubMed searches to 22 June 2020 using the search terms ‘COVID‐19’ or ‘SARS‐CoV‐2’, and ‘immune checkpoint’, ‘nivolumab’, ‘ipilimumab’, ‘pembrolizumab’, ‘avelumab’, ‘cemiplimab’ or ‘atezolizumab’. Only articles in English were included for further analysis.

We identified seven case reports and one case series of patients treated with immune checkpoint inhibitors who developed SARS‐CoV‐2 infections (Table 1), a total of 10 patients. An additional case of coronavirus HKU1 was reported.

Table 1

Details of the patients on immune checkpoint inhibitors with COVID‐19 infection found in the literature.

Author	Serzan et al., 20201	Lovly et al., 20202	Szabados et al., 20203				Schmidle et al, 20204	Artigas et al., 20205	Bonomi et al., 20206	Yekedüz et al., 20207	O'Kelly et al., 20208
Article type	Case report	Case report	Case series				Case report	Case report	Case report	Case report	Case report
Age, years	65	56	52	68	66	72	47	51	65	75	22
Sex	Male	Male	Male	Male	Male	Male	Female	Male	Male	Female	Female
Ethnicity	Caucasian	NS	NS	NS	NS	NS	NS	NS	NS	NS	Not stated
Comorbidities		Diabetes, COPD	Hypertension	Hypertension	Hypertension	Hypertension, diabetes	NS	NS	COPD	COPD, IHD/AF, hypertension, diabetes	Previous chemotherapy, brentuximab
Smoker	NS	Yes	No	Yes	No	Yes	NS	NS	NS	NS	NS
Cancer type	Stage IV melanoma	Small cell lung cancer	Renal cell carcinoma	Renal cell carcinoma	Urothelial carcinoma	Urothelial carcinoma	Stage IV melanoma adj.	Renal cell carcinoma	Lung adenocarcinoma	Stage IV melanoma	Hodgkin lymphoma
I‐C treatment	Ipilimumab 3 mg/kg, nivolumab 1 mg/kg	Carboplatin, etoposide, atezolizumab	Ipilimumab, nivolumab	Ipilimumab, nivolumab	Atezolizumab	Atezolizumab	Nivolumab	Nivolumab	Nivolumab	Nivolumab	Pembrolizumab
Time to symptoms	48 h	48 h	2 cycles	1 cycle	6 months	4 months	7 months	4 months	7 months	27 cycles	6 cycles (every 6 weeks)
Symptoms	Dyspnoea; yellow sputum	Dyspnoea; chest pain cough	Steroids for irAE (rash) after 32 days; dyspnoea; fever; cough	Fever; cough	Dyspnoea. cough; pneumonitis treated with steroids	Cough; diarrhoea; renal failure	Cough; headache; fever; sore throat	Fatigue	Confusion; fever; dyspnoea	Diarrhoea; dyspnoea; fever	Cough; fever; sore throat
CT findings	Ground‐glass infiltrates	Bilateral milk glass infiltrates	Bilateral lung infiltrates	NS	Fibrotic lung changes	NS	Unremarkable	Ground‐glass opacities	CXR: interstitial changes	Bilateral pleural thickening	Bilateral infiltrates
SARS‐Cov‐2 PCR positive	Coronavirus HKU1‐positive	Positive	NS	Positive	Positive	Positive	Positive	Positive	Positive	Positive	Positive
Serology		IgM/IgG SARS‐Cov‐2 antibodies					IgG SARS‐Cov‐2 Antibodies
		Weakly positive					Positive
Treatment	High‐dose corticosteroids, tapered when swab result was positive; nivolumab monotherapy initiated	Methylprednisolone 1 g/day for 2 days; prednisolone 1 mg/kg; prednisolone 2 mg/kg; infliximab 5 mg/mg; vancomycin; piperacillin/tazobactam; immunoglobulins 1 g/kg; steroids weaned; mechanical ventilation; hypoxaemic respiratory; failure; PD‐L1 increased in alveolar walls where viral DNA was detected; IgG response suggested infection before immunotherapy and chemotherapy	High‐flow oxygen; co‐amoxiclav; clarithromycin	Self‐isolation	Self‐isolation	Volume replacement; PIP/TAZ	Supportive	PIP/TAZ; HCQ	Oxygen; antibiotics	Oxygen; PIP/TAZ; oseltamivir; clarithromycin; metronidazole; azithromycin; HCQ. When CT showed ground‐glass opacities, favipravir added	Oxygen; PIP/TAZ; doxycycline; lopinavir; ritonavir; HCQ; azithromycin; lopinavir/ritonavir stopped
ICI treatment	Nivolumab resumed then	IT discontinued	IT not yet recommenced	IT recommenced	IT recommenced	IT recommenced	NS	NS	NA	NA	NS
	discontinued after CR
Outcome	Alive	Dead	Alive	Alive	Alive	Alive	Alive	Not specified	Dead	Dead	Alive

Of the 10 patients with SARS‐CoV‐2, 30% were women and age range was 22–75 years. Half (50%) of the cases had an underlying urological tumour, 20% had metastatic melanoma, 20% had lung cancer and 10% had a haematological malignancy. Regarding treatment, 30% of the patients had received an anti‐PD‐L1 treatment (atezolizumab), 20% a combined anti‐CTLA‐4/anti‐PD‐1 treatment, 40% were treated with nivolumab (anti‐PD‐1) monotherapy and one patient (10%) received pembrolizumab (also anti‐PD1). The effect of comorbidity, smoking status and ethnicity was difficult to ascertain as these were inconsistently recorded. Time from initiation of immune checkpoint inhibitor to the development of COVID‐19 symptoms ranged from 48 h to > 1 year. The treatments for COVID‐19 infection varied considerably, but 70% of cases received antibiotics, 20% antiviral medication and 30% received hydroxychloroquine (some patients received > 1 treatment). Three patients did not require specific therapy. The patient with coronavirus HKU1 received systemic corticosteroids for presumed checkpoint‐mediated pneumonitis. In fact, the clinical and radiological presentation of immune checkpoint‐related pneumonitis may be indistinguishable from that of SARS‐CoV‐2, making early SARS‐CoV‐2 PCR testing crucial. Three patients (30%) died due to coronavirus infection. Of the remaining patients, immune checkpoint therapy was recommenced or planned for four.

We found that only 10 patients with COVID‐19 infection during immune checkpoint inhibition therapy have been reported. However, it is worth noting that 30% of the cases had a very mild clinical course and did not require hospitalization. Moreover, immune checkpoint therapy was safely recommenced in several patients. These points are extremely important given the fear and anxieties of patients with cancer regarding COVID‐19 infection, which may lead some patients to unnecessarily delay or interrupt therapy.

Ultimately, the decision on whether to initiate and/or continue immune checkpoint therapy during the coronavirus pandemic must be based on a detailed consideration of several factors, including tumour burden and progression, comorbidities, existing immunosuppression, palliative vs. adjuvant treatment and alternative treatment options, and cannot be generalized. Geographical coronavirus prevalence should also be considered. Irrespective of the final cancer treatment decision, the importance of facial coverings, social distancing, shielding and hand hygiene should also be emphasized.

Moving forward, there seems to be a strong case for a comprehensive and standardized prospective register of COVID infections during immune checkpoint inhibition therapy, at least at the local and national levels. This would provide vital information to determine how checkpoint inhibition influences the course of the disease, enabling clinicians to counsel their patients adequately. Furthermore, in light of the apparent increased mortality in various ethnic groups,, combined with the potential under‐reporting of ethnicity in the published COVID‐19 dermatological literature, a register would ensure that key risk factors are not overlooked. In the absence of this information, it seems prudent to thoroughly assess all patients due to commence, and those currently undergoing, immune checkpoint therapy, for coronavirus risk factors and symptoms, complemented by early and rigorous SARS‐CoV‐2 PCR testing where clinically indicated and available.