Diagnosis and management of the drug hypersensitivity reactions in Coronavirus disease 19: An EAACI Position Paper.

Coronavirus disease 2019 (COVID-19), a respiratory tract infection caused by a novel human coronavirus, the severe acute respiratory syndrome coronavirus 2, leads to a wide spectrum of clinical manifestations ranging from asymptomatic cases to patients with mild and severe symptoms, with or without pneumonia. Given the huge influence caused by the overwhelming COVID-19 pandemic affecting over three million people worldwide, a wide spectrum of drugs is considered for the treatment in the concept of repurposing and off-label use. There is no knowledge about the diagnosis and clinical management of the drug hypersensitivity reactions that can potentially occur during the disease. This review brings together all the published information about the diagnosis and management of drug hypersensitivity reactions due to current and candidate off-label drugs and highlights relevant recommendations. Furthermore, it gathers all the dermatologic manifestations reported during the disease for guiding the clinicians to establish a better differential diagnosis of drug hypersensitivity reactions in the course of the disease.

INTRODUCTION

Coronavirus disease 2019 (COVID‐19) is a respiratory tract infection caused by a novel member of human coronavirus, the severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2). It causes a wide spectrum of clinical manifestations ranging from asymptomatic cases to patients with mild, uncomplicated illness and severe cases, with or without pneumonia. Hospitalization and oxygen support, and admission to an intensive care unit are required in 14% and 5% of the patients, respectively. Gastrointestinal symptoms and positive viral nucleic acid testing on rectal swabs are considered as indicators of infection in digestive system and fecal‐oral transmission of COVID‐19. Moreover, skin symptoms, including exanthems, may appear during the evolution of the disease leading to differential diagnosis with drug hypersensitivity reactions (DHRs).

In critically ill patients, COVID‐19 can be complicated by acute respiratory distress syndrome (ARDS), septic shock, and multi‐organ dysfunction syndrome. In such patients, in response to viral infection, the excessive activation and expansion of T lymphocytes and macrophages lead to an overproduction of cytokines, which causes a cytokine storm and a hyperinflammatory state. , Acute hyperinflammation may activate coagulation cascade and inhibit fibrinolytic reaction, thus promoting thrombosis. Coagulopathy and thrombocytopenia are serious complications which increase the risk of hemorrhage and thrombosis and progress to disseminated intravascular coagulation (DIC).

The periodically updated World Health Organisation interim guidance allows reliable comparison of investigational therapeutic interventions as part of randomized controlled trials, provides recommendations for the management, and forms the basis of many institutional or national protocols. Unfortunately, none of the drugs used for COVID‐19 have been proven to be truly effective yet; besides, no specific antiviral drugs have been approved for COVID‐19 by health authorities. , At the moment, there is no specific treatment for COVID‐19, and standard practice of care focuses on treating the clinical symptoms with supportive care.

In this review, diagnosis and management of DHRs, which are expected to be caused by current or candidate repurposed and off‐label drugs used for COVID‐19 treatment mostly based on prior knowledge, are discussed. , , Drugs in this review are classified into four groups according to their potential roles in different phases of the disease as antiviral drugs, antiviral and/or immunomodulatory drugs used in viral pneumonia; anti‐cytokine and anti‐inflammatory drugs considered during macrophage activation syndrome (MAS) and cytokine storm; anti‐inflammatory drugs in ARDS; and anti‐aggregant and anti‐coagulant drugs in coagulopathy (Figure 1). Information of DHRs due to the use of additional drugs for various purposes can be found in the relevant European Academy of Allergy and Clinical Immunology (EAACI) resources. , , , , , , , ,

Figure 1

Currently investigated drugs in COVID‐19 grouped according to their clinical use

Since emerging recent findings are dynamically changing the clinical interventions, it is expected that the list of drugs determined according to current knowledge may change with upcoming recommendations in future.

SKIN MANIFESTATIONS INDUCED BY COVID‐19

There have been increasing reports of dermatologic manifestations associated with COVID‐19 (Table 1). It is knowledge, although in progress, rapidly evolving as evidenced by most publications being ahead of print and available only in an electronic version or reported in networks.

Table 1

Skin manifestations reported associated with COVID‐19

Manifestation	Clinical description	Relative frequency*	Similarity to skin rashes of other infections	References
1. Skin manifestations similar to those in other viral infections
Acute urticaria	Sudden appearance of wheals with a fleeting nature. Continual appearance and disappearance of new lesions is characteristic.	19%	Unspecific for COVID‐19; infections are common elicitors for acute urticaria	(4, 22, 23, 31)
Maculopapular exanthem (“erythematous rash”)	Acute erupting, widespread distribution of multiple small, round to oval erythematous macules and/or papules with different degrees of confluence. Mostly trunk, low pruritus.	47%	Unspecific for COVID‐19; infections are common elicitors for maculopapular exanthem	(4, 21, 22, 24, 31, 32)
Varicella‐like exanthem (“chickenpox‐like rash”)	Monomorphic papulovesicular skin eruption. Erythematous papules and vesicles bilaterally and symmetrically mostly on the trunk.	9%	May be more specific for COVID‐19, vesicles are quite uncommon for virus exanthems and more specific for varicella	(4, 22, 25, 26)
Symmetrical intertriginous exanthem	Flexural erythematous maculopapular exanthem on axillary lesions and trunk +/‐antecubital fossa.	Individual case reports	Untypical for infectious exanthems	(30)
2. Skin manifestations associated with vascular pathologies
Purpuric exanthem (“purpuric rash”)	Skin rash with petechiae.	Individual case reports	Untypical for infectious exanthems, except, for example Parvovirus B19	(22, 33)
Erythema ab igne (“livedo reticularis”)	Transient macular erythema in a broad reticular pattern on thigh unilaterally.	6% together with cutaneous acro‐ischemia	Untypical for infectious exanthems	(34)
Chilblain‐like lesions	Acute‐onset, violaceous, infiltrated, and painful plaques on the toes and lateral feet. Vesicles and erosions may be present.	19%	Untypical for infectious exanthems	(22, 35, 36, 37)
Cutaneous acro‐ischemia	Finger and toe cyanosis, purpura, hematoma, skin bulla, and dry gangrene.	6% together with Erythema ab igne (“livedo reticularis”)	Typical for severely ill patients with sepsis	(38, 39)

Relative frequency in percent of this skin manifestations associated with COVID‐19 infections according to Ref. (26). In cases, where no numbers are given, only individual case reports do exist.

According to pathogenetic mechanisms, skin manifestations reported so far can be divided into (1) skin manifestations similar to those in other viral infections and (2) skin manifestations related to thrombovascular events and vascular pathologies.

Skin manifestations similar to those in other viral infections

During the COVID‐19 outbreak in China, it was not a focus to document skin manifestations. Consequently, skin rash has only been reported in 2 out of 1.099 infected patients (0.2%). In contrast, a study by dermatologists from Italy reported skin manifestations in 18/88 patients (20.4%) with COVID‐19. Cutaneous manifestations seen were either erythematous rash (n = 14), widespread urticaria (n = 3), or chickenpox‐like vesicular rash (n = 1). In Spain, among 375 patients with suspected or confirmed COVID‐19, maculopapular eruptions (MPEs), sometimes similar to pityriasis rosea, were observed in 47% of the cases, urticarial lesions in 19%, and vesicular eruptions of the trunk in 9%. Another case of urticaria was presented in France (Figure 2A) and patients with morbilliform exanthem in the USA(Figure 2B). Varicella‐like lesions predominantly on the trunk were described in 22 patients with proven COVID‐19 infection in Italy. Predominance of vesicles was reported in 54.5% and generally mild itching in nine (40.9%) patients. The vesiculopapular exanthem appears to develop early in the course of the disease (Figure 2C). , An outbreak of severe Kawasaki‐like disease has been reported at epicenters of COVID‐19 infection also associated with a polymorphic rash in 30%‐50% of affected children. , In one case, picture of a urticaria‐like rash in a 6‐month‐old child with Kawasaki‐like disease associated with COVID‐19 infection is shown (Figure S1). Two patients with bilateral flexural exanthems resembling systemic drug‐related intertriginous exanthems (SDRIFE), one with axillary purpuric lesions associated with thrombocytopenia, have been published (Figure 2D). A prospective study from France reported a prevalence of 5/103 (4.9%) and confirmed association of pruritic erythematous rash (n = 2) and urticaria (n = 2) with COVID‐19 infections ; they additionally observed one oral herpes simplex virus type 1 reactivation. The histopathological picture of exanthematic skin lesions generally resembles that of viral exanthems. However, in individual patients, early microthrombi and an interface dermatitis with necrotic keratinocytes surrounded by lymphocytes have been reported.

Figure 2

Skin manifestations similar to those in other viral infections. A, Urticaria, B, Morbilliform maculopapular exanthem, C, Vesiculopapular (chickenpox‐like) exanthem, D, Intertriginous purpuric rash

Skin manifestations associated with thrombovascular events and vascular pathologies

COVID‐19 exanthems have also been reported with petechiae and low platelet count resembling dengue. In two patients, unilateral lesions on the thigh resembling livedo reticularis or erythema ab igne have been described with microthromboses discussed as possible etiology (Figure 3A).

Figure 3

Skin manifestations associated with thrombovascular events and vascular pathologies. A, Transient unilateral livedo reticularis (erythema ab igne), B, COVID‐19‐induced chilblains, C, Acro‐ischemia with cyanosis, skin bulla, and dry gangrene in critically ill patient

Chilblain‐like skin lesions have been frequently reported to be associated with COVID‐19 , , , (Figure 3B). They appear in up to 19% of patients, typically in mildly affected ones, and late in the evolution of the disease. , Vesicles, pustules, and erosions on these violaceous plaques may occur. In Spain, they were observed in 19% of 375 cases.

Seven patients had cutaneous acro‐ischemia including finger and toe cyanosis, skin bulla, and dry gangrene associated with COVID‐19 infection‐induced hypercoagulation including definitive DIC in four patients. Five of these patients finally died (Figure 3C). A catastrophic microvascular injury syndrome mediated by activation of complement pathways and an associated procoagulant state were described in severe COVID‐19 with purpuric skin rash in 3/5 patients.

In conclusion, the prevalence of cutaneous manifestations in COVID‐19‐infected patients has been reported between 0.2%, 4.9%, and 20.9%. , , Most skin manifestations resemble cutaneous involvement commonly occurring during viral infections, that is, erythematous rash and acute urticaria. Drug exanthems have to be considered as differential diagnosis. Vesicular varicella‐like exanthems may be more specific for COVID‐19. Flexural distribution, and petechiae as well as erythema ab igne‐like lesions have been described. Violaceous, infiltrated painful plaques resembling chilblains have been frequently reported and discussed as typical manifestations. Necrotic lesions occurred in older and in severely ill patients with increased mortality. Cutaneous acro‐ischemic microthromboses and small blood vessel occlusion have to be further explored for their causality and specificity for COVID‐19 manifestations.

ANTIVIRAL AGENTS USED FOR VIRAL PNEUMONIA

Clinical use in COVID‐19

Most antiviral agents used for COVID‐19 act either by inhibiting RNA‐dependent RNA polymerase [remdesivir (GS‐5734)] or proteases [lopinavir/ritonavir (LPV/r), favipiravir (FPV), ribavirin, and darunavir]. , , , Additionally, umifenovir plays a role in viral entry by inhibiting the hemagglutinin‐mediated membrane fusion, and oseltamivir is a neuraminidase inhibitor which blocks the release of viral particles from the host cells in influenza infection. Remdesivir and FPV are considered to be the most effective agents and are mostly used in combination with other COVID‐19 medications like hydroxychloroquine. , , , Oseltamivir is recommended for concomitant influenza infection. Darunavir or LPV/r can be concomitantly administered with chloroquine or hydroxychloroquine.

Hypersensitivity reactions

Drug hypersensitivity reactions to ribavirin, darunavir, LPV/r, remdesivir, and oseltamivir are rarely reported, whereas no DHRs to favipiravir and umifenovir are known at present , , , , , (Table 2).

Table 2

Hypersensitivity reactions due to drugs with antiviral properties investigated for the treatment of COVID‐19 in clinical trials or in vitro studies

Drug groups	Drugs	Purpose of use in COVID‐19	Hypersensitivity reactions	In vivo tests in IHRs	In vivo tests in NIHRs	In vitro tests for IHRs	In vitro tests for NIHRs	Desensitization
Antiviral drugs	Favipiravir	Viral pneumonia	None
	Lopinavir/Ritonavir		AGEP 61 , MPE 60
	Darunavir/Ritonavir		MPE 66 Vesiculobullous lesions 48 , 64					DNIHR 66 , 67
	Umifenovir (Arbidol)		None
	Ribavirin		Pruritus 46 , 55 Eczema 46 , 53 , 54 , 55 Urticaria 58 MPE 56		DPT 56		LTT 57	DNIHR 58 , 59
	Remdesivir (GS‐5734)		MPE 51
	Oseltamivir		Anaphylaxis 68 SJS/TEN 46 , TEN 47	SPT 68			LTT 50
Immunomodulatory drugs	Azithromycine		MPE 73 , 74 , ACD 75 , 76 , FDE 77 AGEP 78 , DRESS 79 SJS 80 , 81 Anaphylaxis 72 Urticaria 71 Leukocytoclastic vasculitis 82 Hypersensitivity myocarditis 83	SPT 84 , 85 IDT 84 , 85 DPT 87	PT 75 DPT 87			DIHR 88
	Hydroxychloroquine / Chloroquine		MPE 90 , AGEP 92 , 103 DRESS 93 , 94 , 95 Erythema multiforme 96 Bullous erythema 97 SJS/TEN 99 , 100 Photoallergic dermatitis 101 ACD 102 Anaphylaxis 109 , 110	SPT 109	PT 96 , 98 , 103 DPT 90			DNIHR 104 , 105 , 106 , 107 , 108 DIHR 109 , 110
	Auranofin		None
	Interferons		Local reaction 116 Urticaria 119 , 122 , 123 Eczema 119 , FDE 120 , MPE 119 Anaphylaxis 124 , 125	SPT 122 , 123 , 124 DPT 125	IDT 119 , 126 PT 119			DIHR 123 DNIHR 119 , 127
	Nitazoxanide		None
	Ivermectin		FDE 129 , DRESS 130 , SJS 131 , TEN 132

Abbreviations: ACD, Acute contact dermatitis; AGEP, Acute generalized exanthematous pustulosis; BAT, Basophil activation test; DIHR, Desensitization for immediate hypersensitivity reactions; DNIHR, Desensitization for nonimmediate hypersensitivity reactions;DPT, Drug provocation test; DRESS, Drug‐related eosinophilia systemic symptoms; FDE, Fixed drug eruption; IDT, Intradermal test; IHR, Immediate hypersensitivity reaction; LTT, Lymphocyte transformation test; MPE, Maculopapular eruption; NIHR, Nonimmediate hypersensitivity reaction; PT, Patch test; SJS, Stevens Johnson syndrome; SPT, Skin prick test; ST, Skin test; TEN, Toxic epidermal necrolysis.

MPE

Vesiculobullous lesions ,

Pruritus ,

Eczema , , ,

Urticaria

MPE

Anaphylaxis

SJS/TEN , TEN

MPE , , ACD , , FDE

AGEP , DRESS

SJS ,

Anaphylaxis

Urticaria

Leukocytoclastic vasculitis

Hypersensitivity myocarditis

SPT ,

IDT ,

DPT

PT

DPT

DIHR

MPE , AGEP ,

DRESS , ,

Erythema multiforme

Bullous erythema

SJS/TEN ,

Photoallergic dermatitis

ACD

Anaphylaxis ,

PT , ,

DPT

DNIHR , , , ,

DIHR ,

Local reaction

Urticaria , ,

Eczema , FDE , MPE

Anaphylaxis ,

SPT , ,

DPT

IDT ,

PT

DIHR

DNIHR ,

Ribavirin

Ribavirin is used in combination with pegylated‐interferon α2a (peg‐IFN‐α2a) for treating chronic hepatitis C, and both have been associated with several cutaneous DHRs. Ribavirin alone causes dermatitis, alopecia, and photoallergic eczematous reactions, , and the risk of DHR increases with combination therapy: rash [response rate (RR), 1.74; 95% confidence interval (CI), 1.17‐2.6], dermatitis (RR, 1.67; 95% CI, 1.21‐2.30), and pruritus (RR, 1.62; 95% CI, 1.29‐2.02). A meta‐analysis revealed that, on combination therapy, mild to moderate cutaneous reactions appear in 13.3% of patients, localized cutaneous reactions in2.6%, generalized reactions‐pruritus, skin xerosis, and eczematous changes in 10.3%, alopecia in 4.1%, and exacerbation of lichen planus in less than 1% (Table 2).

The etiological diagnosis is difficult in case of combination therapy. A drug provocation test (DPT) confirmed the diagnosis of ribavirin hypersensitivity in a patient having MPE due to combined use of peg‐IFN‐α2a and ribavirin. In another case, an erythema multiforme‐type drug eruption occurred with peg‐IFN‐α2a, ribavirin, and/or fluvastatin sodium therapy and a positive lymphocyte transformation test (LTT) confirmed the diagnosis of ribavirin hypersensitivity. Successful desensitization protocols were reported , (Table 2).

Lopinavir/ritonavir (LPV/r)

Lopinavir/ritonavir, either alone or in combination, has been rarely reported to be associated with DHRs. In human immunodeficiency virus (HIV)‐infected patients who received LPV/r combination, MPE rate was reported as 2%‐4%. Acute generalized exanthematous pustulosis (AGEP) was described in two cases receiving LPV/r (Table 2).

In a multicentre randomized study that evaluated the long‐term efficacy and safety of the combination of efavirenz or LPV/r plus abacavir/lamivudine, 2/63 patients in the LPV/r group discontinued the study because of a DHR.

In a recent cohort of 199 severe COVID‐19‐infected patients who received LPV/r combination, only two (1%) experienced self‐limited skin eruptions. A recent study evaluating 217 patients from China revealed that most of the adverse drug reactions (ADRs) were associated with LPV/r and umifenovir with 63.8% and 18.1%, respectively, and history of a drug allergy was higher in these patients (8.5%) comparing with the ones without ADRs (2.2% vs, P < .044).

Darunavir

Darunavir can induce a variety of delayed skin eruptions from mild MPE in most cases, to severe bullous cutaneous reactions in HIV‐infected patients. , A phase III randomized clinical trial performed in 604 patients treated with darunavir/r or LPV/r showed that the percentage of patients experiencing rash was higher in those receiving darunavir/r compared with others (16% vs 7%). Two patients receiving darunavir/r required treatment cessation due to a severe rash. Darunavir contains a sulfonamide moiety and should be used with caution in patients with a known sulfonamide allergy. Desensitization was reported to be successful in patients with nonimmediate hypersensitivity reactions (NIHRs) to darunavir , (Table 2).

Oseltamivir

Oseltamivir, used in influenza, causes rare hypersensitivity reactions although close monitoring of patients is important as two cases with Stevens‐Johnson syndrome (SJS)/Toxic epidermal necrolysis (TEN) have been reported, , with only one being confirmed by LTT. Another case report revealed anaphylaxis due to oseltamivir confirmed by a skin prick test (SPT) (Table 2).

Remdesivir

A recent multicentre study showed that only one (1.6%) out of 61 patients with COVID‐19, experienced MPE during remdesivir treatment and therefore discontinued it prematurely (Table 2).

ANTIVIRAL AND/OR IMMUNOMODULATORY DRUGS USED FOR VIRAL PNEUMONIA

Azithromycin

Clinical use in COVID‐19

Azithromycin interferes with virus internalization process in influenza infection and has shown clinical effects in COVID‐19‐infected patients, although its mechanism against SARS‐CoV‐2 remains unclear.

Hypersensitivity reactions

Regarding immediate hypersensitivity reactions (IHRs), urticaria is the most frequent manifestation ; furthermore, anaphylaxis can occur. Concerning NIHRs, MPE is described to occur independently or only in the presence of a concurrent infection. Azithromycin has been implicated in contact dermatitis in occupational and nonoccupational settings. Cases of fixed drug eruption (FDE), AGEP, and Drug Reaction with Eosinophilia and Systemic Symptoms (DRESS), SJS, , leukocytoclastic vasculitis, and hypersensitivity myocarditis were reported(Table 2).

Diagnosis is complex as skin testing is not validated, presenting discrepancies in nonirritating dilutions for SPT and intradermal test (IDT). , For NIHRs, positive responses to patch tests (PTs)were described. In addition, no validated in vitro tests are available. Oral DPT remains as the gold standard for diagnosis. A successful desensitization protocol was reported in a case of mast cell activation syndrome (Table 2).

Hydroxychloroquine/Chloroquine

Hydroxychloroquine/chloroquine have in vitro antiviral effects against SARS‐Cov‐2 by preventing virus/cell fusion, and immunomodulatory effects by inhibiting production of inflammatory cytokines.

Dermatologic ADRs are difficult to be distinguished as a side effect of or an allergic reaction to these drugs or a flare of the underlying dermatological disease. , The most common manifestation is mild pruritic MPEs within initial 4 weeks of treatment. High association with AGEP [OR: 39 (8‐191)] was described. Cases of DRESS, , pustular DRESS, erythema multiforme, bullous erythema, SJS/TEN, , , photoallergic dermatitis, and occupational contact dermatitis have been reported(Table 2).

PTs are reported to be useful for the diagnosis of NIHRs, , , confirming a T cell–mediated mechanism. However, in a series of 14 patients with ADRs due to chloroquine/hydroxychloroquine, skin tests (STs) were negative in all cases. DPT is useful in nonsevere cutaneous ADRs in order to differentiate allergic reactions from dermatological adverse effects since only 30% of the patients reporting cutaneous ADRs reveal a positive DPT. Successful desensitization protocols of hydroxychloroquine in MPE were reported. , , , Recently, a 5‐hour desensitization protocol for nonimmediate urticaria was successfully administered (Table 2).

Two cases of IHR were reported , and one was confirmed by SPTs ; however, there are no available data for in vitro diagnosis. A hydroxychloroquine desensitization procedure that enables the turning of positive SPTs into negative was published. In a case of anaphylaxis, a 7 day‐desensitization procedure was successfully performed with premedication (Table 2).

Auranofin

Auranofin is an anti‐inflammatory compound that can possibly inhibit the replication of SARS‐CoV‐2 in cell culture and reduce the expression of cytokines caused by SARS‐CoV‐2 and the associated lung damage.

There are no reported hypersensitivity reactions due to auranofin.

Interferons

Type I IFNs (IFN‐α and IFN‐β) can inhibit the replication of both SARS and Middle East respiratory syndrome coronavirus (MERS‐CoV) and are recommended in combined therapies with other antiviral agents. ,

Cutaneous eruptions induced by IFNs are common, with an incidence of 13%‐23%. , Localized reactions at injection sites are most frequent at 48 weeks. Diffuse skin symptoms including urticaria, generalized eczema, papules are common and mostly treated with symptomatic treatment. , , Among 26 patients with nonimmediate reactions to IFNs, 12 cases reported generalized eczema, 10 MPE, 3 generalized urticaria, and 1 lichenoid eruption. Cases of FDEs, and subacute cutaneous lupus were described (Table 2).

There are few case reports of immediate urticaria , and anaphylaxis. , For IHRs to IFN‐β, positive STs were reported. , For NIHRs, PTs have a low value and are not recommended, whereas delayed reading IDTs are useful. , A positive DPT was reported in a patient experiencing anaphylaxis due to peg‐INF‐α2a with negative STs. Successful desensitization protocols both for IHRs and NIHRs , due to different IFNs were reported (Table 2).

Ivermectin

Ivermectin is an antiparasitic drug also shown to have an in vitro activity against SARS‐CoV‐2 by inhibition of viral replication.

Rare case reports of multiple FDEs, confirmed DRESS by skin biopsy and blood eosinophilia, confirmed SJS and TEN by skin biopsy were published (Table 2). No data about STs, in vitro tests, or DPT are available. In addition, no cases of desensitization were reported.

Nitazoxanide

Nitazoxanide is an antiparasitic agent which also has antiviral activities. Combined with hydroxychloroquine or azithromycin, a synergistic effect has been suggested as hydroxychloroquine and azithromycin inhibit viral entry and fusion, while nitazoxanide upregulates innate immune response to prevent ongoing viral replication in COVID‐19.

No DHRs to nitazoxanide are reported.

ANTI‐CYTOKINE/ANTI‐INFLAMMATORY DRUGS USED FOR MAS/CYTOKINE STORM/ARDS

Tocilizumab

Tocilizumab, an anti‐IL‐6 receptor humanized monoclonal antibody, is under investigation for treatment of COVID‐19 and has shown promising results in cytokine storm.

The rate of all ADRs to tocilizumab is reported to be around 8%, among them 0.1%‐0.7% are DHRs. DHRs to tocilizumab are both NIHRs , and IHRs , , , (Table 3). In an adult cohort, the incidence of IHRs was reported as 5.5% whereas in a pediatric cohort it was 13.6%.

Table 3

Hypersensitivity reactions due to other drugs investigated for the treatment of COVID‐19–related complications in clinical trials or in vitro studies

Drug groups	Drugs	Purpose of use in COVID‐19	Hypersensitivity reactions	In vivo tests in IHRs	In vivo tests in NIHRs	In vitro tests for IHRs	Desensitization
Anti‐cytokine or anti‐inflammatory drugs	Tocilizumab	Cytokine storm/MAS	Papular skin lesions 136 Nonimmediate urticaria 141 Anaphylaxis 137 , 139 DRESS 142 , 143 AGEP 145 , SJS 144	SPT 137 , 139 IDT 137 , 139 DPT 137 , 139	IDT 140		DIHR 134 DNIHR 141
	Sarilumab		Pruritic rash 155
	Anakinra		ISR 147 , 148 U/Angioedema 149 Anaphylaxis 150 , 151 Erythematous plaques 152	SPT 150 , 151 , IDT 150 , 151			DIHR 149 , 151 DNIHR 152
	Canakinumab		U 137
	JAK inhibitors Baricitinib		Palmoplantar pustulosis 160
	JAK inhibitors Ruxolitinib		morbiliform rash, exfoliative dermatitis 159
	JAK inhibitors Tofacitinib		U 161 , palmoplantar pustulosis 162
	Cyclosporine		Anaphylaxis 164 , 165 , 166 , 167	SPT 164 , 166 IDT 164 , 166		BAT 166
Anti‐inflammatory drugs	Glucocorticoids	Cytokine storm/MAS ARDS	IHR 178 , 179 , 180 , 181 , 182 , 183 NIHR 178 , 179 ACD 179	ST 178 , 179 , 180 , 181 , 182 , 183 , 184 , 185 DPT 180 , 181 , 182 , 183 , 184 , 185	IDT 179 , 189 PT 179 , 189 DPT 179 , 189	sIgE 181 , 182 BAT 181 , 182	DIHR 186 , 187
	Colchicine		Anaphylaxis 151 FDE 168 , MPE 169		DPT 168 PT 170		DNIHR 169
	Eculizumab		IHR 172 , 173 , 174 Anaphylaxis 174	SPT 174 , IDT 174			DIHR 174
Anti‐coagulant or anti‐aggregant drugs	Heparin Enoxaparin	Coagulopathy	ISR 192 GDE 195 , 196 DRESS 198 , SJS 199 HIT 200 , 201 , 202 IHR 197 , 204	SPT 17 DPT 191 , 192	PT 17 DPT 191 , 192 IDT 17	BAT 204 , 206 , 207	DIHR 209 , 210
	Dipyridamole		Eczema 215 Anaphylaxis 216 , 217		PT 215

Abbreviations: ACD, Acute contact dermatitis; AGEP, Acute generalized exanthematous pustulosis; BAT, Basophil activation n test; DIHR, Desensitization for immediate hypersensitivity reactions; DNIHR, Desensitization for nonimmediate hypersensitivity reactions; DPT, Drug provocation test; DRESS, Drug related eosinophilia systemic symptoms; FDE, Fixed drug eruption; HIT, Heparin induced thrombocytopenia; IDT, Intradermal test; IHR, Immediate hypersensitivity reaction; ISR, Injection site reaction; GDE, Generalized delayed exanthema; LTT, Lymphocyte transformation test; MPE, Maculopapular eruption; NIHR, Nonimmediate hypersensitivity reaction; PT, Patch test; SJS, Stevens Johnson syndrome; SPT, Skin prick test; ST, Skin test; TEN, Toxic epidermal necrolysis; U, Urticaria.

Papular skin lesions

Nonimmediate urticaria

Anaphylaxis ,

DRESS ,

AGEP , SJS

SPT ,

IDT ,

DPT ,

DIHR

DNIHR

ISR ,

U/Angioedema

Anaphylaxis ,

Erythematous plaques

SPT , , IDT ,

DIHR ,

DNIHR

U ,

palmoplantar pustulosis

SPT ,

IDT ,

IHR , , , , ,

NIHR ,

ACD

ST , , , , , , ,

DPT , , , , ,

IDT ,

PT , DPT ,

sIgE ,

BAT ,

DIHR ,

Anaphylaxis

FDE , MPE

DPT

PT

IHR , ,

Anaphylaxis

ISR

GDE ,

DRESS , SJS

HIT , ,

IHR ,

SPT

DPT ,

PT

DPT ,

IDT

Eczema

Anaphylaxis ,

Regarding NIHRs, cases of nonimmediate urticaria, DRESS, , SJS, and AGEP were reported. Younger age, shorter stature, lighter weight, and increased disease activity in the early period of tocilizumab administration have been identified as risk factors for DHRs.

Although not standardized, DPTs, SPTs, and IDTs were used for diagnosis of IHRs in case reports. , Only one study revealed that STs have a low negative predictive value in NIHR. Desensitization to tocilizumab in NIHRs was effectively applied in a weekly scheme with premedication in one case. Rapid drug desensitization is successfully and routinely used for IHRs , (Table 3).

Anakinra

Anakinra, a recombinant IL‐1 receptor antagonist, is under investigation for the treatment of cytokine storm seen during COVID‐19.

Anakinra causes ADRs in 75% of patients. Many of them are related to injection site reactions within the first weeks of application and can present either as an IHR or NIHR. , Systemic IHRs such as urticaria, angioedema, anaphylaxis, , , and NIHRs as infiltrating erythematous skin plaques were rarely reported as single cases. IHR after a first dose of anakinra was reported in a case possibly due to components that are able to induce a direct mast cell degranulation , (Table 3).

For evaluating IHRs to anakinra, SPTs and IDTs were performed with the undiluted drug. , For both IHRs , and NIHRs, successful desensitization protocols were reported (Table 3).

Sarilumab

Sarilumab, another IL‐6 receptor antagonist, is under investigation in a phase II/III clinical trial in patients with severe COVID‐19 infection.

It is generally a well‐tolerated drug; however, it can cause local reactions on injection site. In an open‐label study, in 3% of the patients it caused a pruritic generalized rash which did not affect the treatment (Table 3).

Canakinumab

Canakinumab, a high‐affinity human anti‐IL‐1β monoclonal antibody, is considered as a candidate in treatment of severe COVID‐19.

This anti‐IL‐1 agent is normally well tolerated and indicated as an alternative in cases with an anaphylactic reaction to anakinra. However, there is a recently reported case who developed immediate diffuse urticaria after the tenth canakinumab administration and was prevented from further reactions with cetirizine premedication (Table 3).

Janus kinase (JAK) inhibitors (Baricitinib, Ruxolitinib, Tofacitinib)

Janus kinase inhibitors are under investigation for their potential role in regulating the overactive signaling in the JAK‐STAT pathway seen during cytokine storm in critically ill COVID‐19‐infected patients. Baricitinib with its potential to inhibit clathrin‐mediated endocytosis, and its ability to ameliorate associated chronic inflammation in interferonopathies is expected to show promising results in ongoing clinical trials of COVID‐19. ,

Few cases were reported: one with a morbiliform eruption and exfoliative dermatitis due to ruxolitinib, another one with palmoplantar pustulosis due to baricitinib, and cases of acute urticaria and palmoplantar pustulosis due to tofacitinib (Table 3).

Cyclosporine

Cyclosporine A prevents the transcription of genes encoding cytokines like IL‐2 and inhibits the replication of diverse coronaviruses at noncytotoxic, low‐micromolar concentrations in vitro.

Rare cases of pruritus, urticaria, angioedema, and anaphylaxis were reported. , , The possible mechanisms can be both immunologic and nonimmunologic, which seems to depend on the administration route and formulation. In some cases, DHRs have been attributed to the additives such as castor oil or Cremophor EL. SPTs and IDTs or basophil activation test (BAT) can be used for the diagnosis of cyclosporine‐ and additive‐induced IgE‐mediated IHRs , , (Table 3).

Colchicine

It is a nonselective inhibitor of NLRP3 inflammasome which is thought to be a major pathophysiologic component of ARDS and/or acute lung injury seen in COVID‐19.

Rare cases of anaphylaxis, confirmed FDE with DPT and successfully desensitized MPE were reported. For PTs, it is recommended to dilute colchicine to 1% in petrolatum (Table 3).

Eculizumab

Eculizumab, a humanized anti‐C5 monoclonal antibody, is under investigation as a candidate drug to play a role in the thrombotic microvascular injury mediated by complement activation causing lung injury either due to severe pneumonia or ARDS in severe COVID‐19. , ,

Immediate hypersensitivity reactions or infusion reactions due to eculizumab are very rare. , A case of anaphylaxis diagnosed with STs was successfully desensitized with a rapid protocol (Table 3).

Glucocorticoids

In COVID‐19‐infected patients, the use of glucocorticoids (GCs) is rather controversial. , Early start of GCs could be helpful for patients who have an overly exuberant inflammatory response or are at high risk of developing ARDS, whereas the benefit of GCs as rescue treatment remains doubtful.

IHRs to GCs are overall rare and mostly IgE‐mediated. , , , , , In a review of the literature from 2004 to 2014, anaphylaxis was the most common manifestation reported (60.8%, 73/120 reactions) followed by urticaria and/or angioedema (26.7%). Methylprednisolone was implicated in 41% of reactions, followed by prednisolone (20%), triamcinolone (14%), and hydrocortisone (10%).

In most subjects with IHRs, it is possible to identify the culprit and safe alternative GCs by performing immediate‐reading STs. , , , , , , , In the aforementioned review, 74.1% of 112 STs carried out with GCs suspected of being responsible for reactions were positive. In some subjects, positive STs were associated with positive serum‐specific IgE assays and BATs , (Table 3).

Immediate hypersensitivity reactions to medication components other than the GC itself, such as succinate ester used to enhance the solubility in parenteral preparations, have been described. , Hence, when evaluating a reaction to an esterified GC, it is advisable to include in STs the suspected GC and the same GC without the ester component, or with a different ester.

Immediate hypersensitivity reactions to excipients or preservatives in GC preparations, such as lactose, carboxymethylcellulose, polyethylene glycol, and hexylene glycol, have also been reported. , Therefore, testing should be performed with a preservative free GC, in addition to preservative testing per se if needed (Table 3). A study proposed a comprehensive diagnostic algorithm to evaluate hypersensitivity reactions to GCs, as well as to their components and preservatives. This algorithm included STs with Carmellose® eye drops in subjects who had reacted to carboxymethylcellulose‐containing GCs and with cow's milk proteins in those who had reacted to lactose‐containing GCs.

In the allergy workup, negative results in STs should be confirmed with DPTs. , , , , , DPTs are also recommended to ensure tolerance of alternative preparations. Cross‐reactivity patterns based on structural characteristics have not been clearly established for IHRs as they have been for allergic contact dermatitis. DPTs have shown that patients often tolerate alternative GCs belonging to the same chemical group as the responsible GC. , Desensitization to methylprednisolone has been successfully performed , (Table 3).

Nonimmediate hypersensitivity reactions following systemic administration of GCs have been more rarely reported than IHRs; most reports concerned isolated cases of eczematous or exanthematous skin eruptions , (Table 3). Some are systemic contact dermatitis, occurring in patients with previous contact dermatitis to GCs. They can be revealed by a Baboon syndrome, characterized by a buttock erythema associated to a symmetric, flexural erythema.

Most patients do not have a previous topical sensitization. In NIHRs, the main feature is MPE, but other clinical aspects can also occur such as annular erythema, erythroderma, SDRIFE, AGEP, FDE, and a few cases of SJS (Table 3).

Nonimmediate hypersensitivity reactions can be T cell–mediated, and PTs, together with delayed‐reading IDTs, are useful tools for evaluating them. PTs have to be read at 2, 4, and also 7 days. Even though delayed reading IDTs are more sensitive than PTs, the sensitivity of the former is limited. Therefore, DPTs are often necessary to diagnose NIHRs. In a study by Padial et al, only 2 of the 38 patients with NIHRs to GCs displayed positive delayed‐reading IDTs and PTs to the responsible GCs (ie, dexamethasone and betamethasone), while 21 of the 32 negative patients who agreed to undergo DPTs reacted to them, experiencing almost exclusively delayed‐appearing urticarial eruptions or MPEs (Table 3).

ANTI‐COAGULANT AND ANTI‐AGGREGANT DRUGS USED FOR COAGULOPATHY

Heparin and low molecular weight heparins (LMWHs)

Heparin [unfractionated heparin (UFH)] and LMWHs are administered for treatment or prophylaxis of thrombosis, and therefore, it is used for the coagulopathy observed during COVID‐19.

Unfractionated heparin may induce all types of DHRs, mostly type IV and type II. Cutaneous NIHRs to subcutaneous heparin occur at the injection site as itchy erythematous or eczematous plaques usually on the 7th‐10th day of treatment, although they can appear on the 1‐3th day in case of antecedent sensitization. Risk factors for NIHRs to heparin are obesity, female gender, old age, pregnancy, and repeated exposures. , If the treatment is continued regardless of a local reaction, the patient may develop generalized eczema or exanthem. , Patients with a NIHR to UFH or LMWH at injection site usually tolerate intravenous administration of UFH. Cross‐reactivity among LMWHs has been reported in NIHRs. However, fondaparinux is generally well tolerated in patients who react to LMWHs. Heparin may induce DRESS and SJS.

Immune‐mediated heparin‐induced thrombocytopenia (HIT) is induced by IgG antibodies against complex of heparin and platelet‐factor 4 tetramers. HIT manifests as a more than 50% decrease in the platelet count in 5‐10 days after the onset of treatment. The risk of HIT is increased exclusively with UFH. Treatment includes the discontinuation of heparin and the introduction of an alternative anti‐coagulant such as argatroban, fondaparinux, danaparoid, or bivalirudin.

The IgE‐mediated reactions to heparin manifesting as urticaria, angioedema, and anaphylaxis are rare. , , Positive STs with UFH and LMWHs have been reported , , , (Table 3). Cross‐reactivity in IHRs has been reported between UFH and LMWH and among LMWHs.

For IHRs with heparins, diagnostic approach primarily consists of SPTs and IDTs. The results of BAT with UFH and LMWH are controversial. , , Heparin itself may cause a release of histamine, leading to a false‐positive ST. Further serial dilutions of heparin (1:100, 1:1000, 1:10 000) might be needed. IDTs and PTs with the culprit and alternative heparin are performed in NIHRs. PTs, with tape stripping, are less sensitive but may be positive (Table 3).

Drug provocation test is considered when the diagnosis is obscure, tissue pathology is unavailable, or an alternative anti‐coagulant needs to be determined. Subcutaneous DPTs with UFC and LMWHs are performed with increasing doses reaching up to a daily dose on the first day, then are evaluated on three consecutive days and day 7 in case of NIHRs. Intravenous DPTs with UFC may also be necessary to prove tolerance for emergency situations both for IHRs and NIHRs. , A standard protocol for UFH desensitization has not been established yet and published as case reports , (Table 3).

Dipyridamole

Dipyridamole is an inhibitor of phosphodiesterase 3 and 5; thereby, it increases intracellular cAMP and/or cGMP in platelets and inhibits platelet aggregation. Besides, it has antiviral features against several viruses. , Dipyridamole as an adjunctive therapy was demonstrated to be associated with decreased D‐dimer levels in COVID‐19.

Drug hypersensitivity reactions related to dipyridamole are extremely rare. An adult patient with delayed eczematous lesions revealed positive PT results. Anaphylaxis or anaphylaxis like reactions were described in two cases; however, they lack diagnostic tests , (Table 3).

DIAGNOSIS, DIFFERENTIAL DIAGNOSIS, AND MANAGEMENT OF DHRS DUE TO DRUGS INVESTIGATED FOR THE TREATMENT OF COVID‐19

Considering the severity of the disease and the emergent need for interventions, it is important to give accurate and quick diagnostic and therapeutic decisions in case of DHRs during COVID‐19 treatment. However, it is challenging considering the diverse spectrum of drugs introduced either for direct treatment of the disease or other accompanying conditions during the course of the disease especially in severe cases when the disease is prolonged. Consequently, multiple medications applied at a time make a clear‐cut association with one medication more difficult. Furthermore, disease‐related eruptions as an important reason of differential diagnosis can make the diagnosis even harder, considering that the majority of the drugs used are more associated with drug‐related cutaneous NIHRs.

Given the critical state of the disease, the diagnosis can mostly rely on clinical observations without performing in vivo tests which have possible contamination risks and time‐consuming in vitro tests. During a DHR, STs cannot be applied considering the possibility of aggravation and the low diagnostic accuracy expected during ongoing treatment with antihistamines and corticosteroids. When introducing an alternative drug, a DPT based on established methods may be preferred in order to reduce the risk of a possible DHR.

If alternative drugs are not available and underlying DHR is not severe, we can recommend that drugs can be applied with published or tailored desensitization protocols. , When mild, self‐limiting DHR is considered, “treating through” concept, the continued administration of a drug despite a suspected allergic hypersensitivity reaction, can be considered under strict surveillance measures. Our recommendations for the diagnosis and management of DHRs due to drugs administered during COVID‐19 are listed in Box 1.

No equivalent alternatives for the currently off‐label repurposed drugs or novel drugs used in COVID‐19 do exist.

We should extrapolate our knowledge on DHRs from other clinical situations to COVID‐19 considering the scarce experience for the DHRs during the disease.

Various drugs being used in different phases of the disease seem to cause rare but potentially severe DHRs, mostly nonimmediate cutaneous hypersensitivity reactions based on data from limited number of case reports.

The most important differential diagnosis of these DHRs is disease‐related exanthems, which can further be classified into the ones similar to those in other viral infections and the others related to thrombovascular events and vascular pathologies seen during COVID‐19.

Experience of diagnostic and management methods for DHRs due to the drugs used in COVID‐19 depend mostly on few case reports or series.

Knowledge of DHRs is urgently needed from pharmacovigilance registries and data from ongoing clinical trials ofCOVID‐19.

Quick diagnostic and therapeutic decisions in case of DHRs during COVID‐19 are mandatory.

Clinical diagnosis of DHRs during COVID‐19 might mostly rely on clinical observations and basic laboratory findings regarding the need of urgent treatment of COVID‐19.

If the risks of a DHR outweigh the benefits obtained from the drug administration, the offending drug should be discontinued.

When introducing an alternative drug, a DPT may be preferred in order to reduce the risk of a possible DHR.

If an alternative drug cannot be replaced, the offending drug can be administered via desensitization with published or tailored protocols when there are no contraindications.

“Treating through” concept, the continued administration of a drug despite a suspected allergic hypersensitivity reaction, can also be considered under strict surveillance measures if the underlying DHR is mild and self‐limiting, and an alternative drug does not exist.

CONCLUSION

This review brings together all the published information about DHRs due to current and candidate off‐label drugs to treat COVID‐19. The current knowledge depends mostly on previous clinical experience and few published studies or case reports. Hopefully, published literature reveals that most of these drugs rarely cause DHRs but severe reactions may also occur. One limitation of this review is that it includes extremely low number of reports of ADRs seen so far during COVID19 treatment. In near future, we need to obtain data about DHRs during the disease from ongoing clinical trials and DHR registries. Additionally as time passes, we will observe if SARS‐CoV‐2 can aggravate T cell–mediated reactions as some viruses do, and if the hyperinflammation observed during the course of the disease may influence DHRs.

This review also highlights the presence of two different groups of disease‐related exanthems as an important cause of differential diagnosis of DHRs expected during the treatment of the disease. We think that it is extremely important to distinguish these disease‐related eruptions from true DHR‐related skin manifestations considering that the majority of the drugs used are more associated with drug‐related nonimmediate skin reactions.

In near future, further data from ongoing clinical studies and registries established in different countries will enlighten the obscure parts of our understanding on DHRs due to the drugs used in the treatment of COVID‐19 and will possibly enable us to establish accurate diagnostic and treatment approaches for these reactions.

CONFLICT OF INTEREST

None of the authors declare conflict of interest.

Figure S1

Click here for additional data file.