HIV and SJS/TEN
Enhanced expression of immune receptors may also be responsible for the highly elevated incidence of SJS/TEN in HIV-infected patients. Indeed, the mechanism underlying SJS has been found to be due to p-i, where investigated [21, 23, 24]. HIV infection increases HLA expression, and, although there are reduced T cells present, these circulating T cells are highly activated. Thus, the opportunity for T cells to react to drugs and stimulate p-i reactions is dramatically increased. This could explain that the incidence of SJS/TEN (in response to various drugs) is around 100-fold increased in HIV-infected compared to non-infected individuals [25]. Some authors have proposed additional factors such as HIV-induced depletion of immunoregulatory cells and increased oxidative stress as favoring the development of DH [5, 51], but this remains controversial [50].
First DH, then virus release
In the above-mentioned examples, the viral infection primes the immune system for enhanced reactivity to drugs and thus clearly precedes the first manifestations of DH. However, there is an important exception to this sequence of events. In DRESS, DH develops first, and viraemia of endogenous herpes viruses (e.g. human herpes virus 6 (HHV6), cytomegalovirus (CMV), Epstein-Barr virus (EBV)) is subsequently detected. This viraemia can be detected already in the acute stage, but most commonly it is found 3-6 weeks later [52-54]. That means the viraemia develops, at least in the vast majority of cases,after cessation of the drugs which induce DRESS (Fig 3). Herpes virus reactivation is a rather common event, and HHV6 is even included in the Japanese definition of DRESS [54, 55]. Some of the herpes viruses are regularly present simultaneously [56], and often the viraemia remains asymptomatic. Severe complications as a result of this viral reactivation appear to be rare and are described mainly during CMV infections (colitis, myocarditis) [57, 58]. Whether the high systemic cytokine levels (e.g., TNFα and G-CSF) in the acute phase of DRESS play a role in CMV reactivation and related complications [58], is doubtful, as CMV reactivation and these complications usually appear much later [59].
An intense p-i stimulation, which involves the activation and expansion of many (polyclonal) CD4+ and CD8+ T cells, underlies DRESS [9, 10, 60]. This stimulation develops over weeks and is mirrored by lymphocytosis with many lymphoblasts in the peripheral blood (part of DRESS definition). These p-i-activated T cells are cytotoxic and infiltrate various tissues [61]. Additionally, DRESS is characterized by a massive activation and expansion of eosinophils in the peripheral blood and target organs.
The p-i stimulation affects both naïve and memory T cells and causes an activation and expansion of various T cells (polyclonal, polyspecific, cytotoxic) [9, 10]. Within this p-i activated T lymphocyte pool, are also T cells that are specific for endogenous herpes viruses. Up to 10% of all CD8+ T cells may be herpes virus-specific [62] and they control viral replication, likely via local IFN-γ release [63][64]. The p-i activation of these T cells may also have a dramatic effect on their effector function. Virus-specific T cells become cytotoxic due to the drug-induced p-i stimulation, and when they encounter their target stimulus (i.e., HHV6, CMV, EBV) in the peripheral tissue, these T cells kill the herpes-infected cells. In fact, an increase in TNFα and IL-6 was observed before HHV6 reactivation in DRESS, which was found to occur 3-4 weeks following the initial diagnosis [65][66]. As a consequence, the endogenous intracellular herpesviruses are released. The release of viral particles results in blood viraemia, most often in the absence of actual viral replication. In many cases, the viraemia remains asymptomatic.
Other authors have linked systemic cytokine storm, as detected in some acute DRESS cases, to viral replication (as shown for TNFα and HHV6) [67, 68], and, as in the case of CMV, serious, potentially fatal complications. However, most viraemia is observed weeks after the cytokine storm.
Beyond viral reactivations, the p-i-activation of T cells in DRESS may also be linked to the occurrence of late autoimmune complications. Instead of virus-specific T cells, the activation and subsequent functional switch to a cytotoxic effector mode may happen in autoreactive T cells [10].
Clinical impact and conclusion
Many factors influence the clinical manifestation of DH in the context of viral infections. These can be virus-, drug- and/or patient-related. Among the virus-related factors are the type and strain of the virus and the length of infection. The drug itself, the dose, its ability to act as a hapten, the patient’s underlying condition, antiviral/antidrug immune response, and potential state of immunosuppression also contribute in this setting. It is challenging, but important for a better understanding of DHR in viral infections, to dissect the respective contribution of these virus-, drug- and/or patient-related factors.
The models proposed can be generalized to various virus infections, and are not dependent on a specific type of virus: The analysis of the timing of symptoms and therapy, as well as the careful monitoring of virus load in SARS-CoV-2-infected patients, has emphasized that DH manifestations are not linked to the presence of viruses, nor that viraemia after DRESS is linked to the presence of a drug. They are due to the virus- or drug-triggered immune stimulations, but are not dependent on the presence of its elicitor (drug or virus). The manifestations of DH can therefore occur after virus clearance – as long as the state of immune activation persists. Vice versa, viral release can occur in the absence of a drug, as illustrated in the late complications of DRESS (Fig 3).
For the clinician, the following questions are important: