5

Thrombosis with Thrombocytopenia Syndrome, Immune Thrombocytopenic Purpura, Capillary Leak Syndrome, and COVID-19 Vaccines

This chapter describes the potential relationship of COVID-19 vaccines and thrombosis with thrombocytopenia syndrome (TTS), immune thrombocytopenic purpura (ITP), and capillary leak syndrome (CLS) (see Boxes 5-1 through 5-3 for all conclusions in this chapter).

THROMBOSIS WITH THROMBOCYTOPENIA SYNDROME

Background

Within months of the introduction of ChAdOx1-S¹ in Europe and the United Kingdom, three reports appeared of an unusual safety signal characterized by the acute onset of unusual thrombotic events and thrombocytopenia 6–24 days after the first dose (Greinacher et al., 2021; Schultz et al., 2021; Scully et al., 2021). The events

___________________

¹ Refers to the COVID-19 vaccine manufactured by Oxford-AstraZeneca.

predominantly affected the cerebral venous sinus circulation and/or the splanchnic venous circulation; less commonly, pulmonary embolism (PE) and other venous beds were involved. Thrombocytopenia was usually significant, frequently in the range of 10–20 × 10⁹ per liter (L), and D-dimer levels were markedly elevated. It was rapidly recognized that these patients had developed antibodies to platelet factor 4 (PF4), which provided the first insights into a potential mechanism for this rare event. Mortality was high if the patients were not treated with a non-heparin anticoagulant and intravenous immunoglobulin (Greinacher et al., 2021; Schultz et al., 2021; Scully et al., 2021). Because of the association between thrombotic events and thrombocytopenia, the syndrome became known as “thrombosis with thrombocytopenia syndrome” or “vaccine-induced immune thrombotic thombocytopenia” (VITT). Diagnostic criteria for VITT have been proposed, with a definite diagnosis consisting of all five of the following criteria: (1) onset of symptoms 5–30 days after vaccination against severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) (or ≤ 42 days in patients with isolated deep vein thrombosis [DVT] or PE); (2) presence of thrombosis; (3) platelet count <150 × 10⁹ per L; (4) D-dimer level >4,000 fibrinogen equivalent units; and (5) positive anti-PF4 antibodies on ELISA (Pavord et al., 2021).

Mechanisms

Similarities in the clinical presentation of TTS to spontaneous, or autoimmune, heparin-induced thrombocytopenia (HIT) quickly led to the recognition that an immune response to PF4 was an important component of TTS. HIT develops in patients within the first 1–2 weeks of therapy with heparin, characterized by thrombocytopenia with or without thrombotic complications. The diagnosis is confirmed by identifying anti-PF4 IgG antibodies that activate platelets in the presence of heparin through binding to the Fcg receptor IIa on platelet surfaces. Epitope mapping studies identified key amino acids on PF4 that form the antibody binding site, which are spatially distinct from its heparin-binding site (Huynh et al., 2019).

Spontaneous HIT was first described in 2008, characterized by an acute presentation with thrombotic complications, thrombocytopenia, and antibodies to PF4 but no prior exposure to heparin (Jay and Warkentin, 2008; Warkentin et al., 2008). As of 2022, fewer than 40 patients with spontaneous HIT had been reported (Warkentin, 2022). Most occurred after an orthopedic surgical procedure with no exposure to heparin, but some developed after an infection. Cerebral venous sinus thrombosis was observed in 6 of 15 patients with spontaneous HIT in nonsurgical settings, a presentation infrequently seen in patients with HIT receiving heparin (Warkentin, 2022). In addition, serum samples from a subset of patients with spontaneous HIT were capable of activating platelets without heparin (Warkentin, 2022).

All three initial reports describing the development of TTS in patients vaccinated with ChAdOx1-S/nCoV-19 noted that with a single exception, anti-PF4 antibodies were present in serum samples (Greinacher et al., 2021; Schultz et al., 2021; Scully et al., 2021). The antibodies could activate platelets, and this effect could be enhanced by the addition of PF4 and blocked by the addition of heparin (Greinacher et al., 2021). In addition, TTS antibodies from some patients were able to activate platelets without heparin (Schultz et al., 2021). Epitope mapping studies found that binding of anti-PF4 antibodies from five patients with TTS after ChAdOx1-S/nCoV-19 was restricted to eight amino acids that were also located within the heparin-binding site (Huynh et al., 2023). Heparin could inhibit binding of the antibodies to PF4, explaining how it could interfere with platelet activation (Singh et al., 2022). Anti-PF4 antibodies from patients with TTS were also associated with excessive thrombus formation containing platelets, neutrophils, and fibrin in a FcgRIIa⁺/hPF4⁺ transgenic mouse model (Leung et al., 2022). Antibodies against PF4 did not cross-react with the SARS-CoV-2 spike protein, indicating that the desired immune response against the virus was not associated with TTS (Greinacher et al., 2021).

Fewer studies have been performed with antibodies from patients diagnosed with TTS after receiving Ad26. COV2.S.² Initial reports confirmed the presence of anti-PF4 antibodies detected by ELISA in most of these patients, but functional testing for anti-PF4 antibodies, using a heparin-dependent serotonin release assay, was frequently negative (See et al., 2021, 2022). In contrast, platelet activation in the presence of PF4 has been shown (Huynh et al., 2023; Kanack and Padmanabhan, 2022), and epitope mapping demonstrated that anti-PF4 antibodies from

___________________

² The COVID-19 vaccine manufactured by Janssen.

patients with TTS after Ad26.COV2.S bind to the same epitopes as anti-PF4 antibodies from patients with TTS after ChAdOx1-S/nCoV-19 (Huynh et al., 2023).

Although considerable data exist characterizing the anti-PF4 antibodies identified in these patients, very little is known about what precipitates the pathological anti-PF4 antibody response in those patients who develop TTS. That almost all cases have been reported in association with one of the two adenovirus vaccines would suggest a class effect. In contrast, the extremely small number of cases reported in patients receiving a messenger ribonucleic acid (mRNA) vaccine is less than the estimated background incidence of cerebral venous sinus thrombosis with thrombocytopenia during the years before COVID-19 (See et al., 2022).

Epidemiological Evidence

Clinical Trial Data

In the Phase 2/3 trial of BNT162b2,³ one recipient presented with DVT, characterized as a nonserious adverse event, 14 days after dose 2, and two did so during the placebo-controlled follow-up period; however, none were associated with thrombocytopenia, according to the manufacturer. No study participants in either treatment group had a clinical manifestation of thrombosis similar to TTS (FDA, 2021a).

In the blinded phase of the trial evaluating mRNA-1273,⁴ eight recipients and six placebo recipients developed DVT, but no events were associated with thrombocytopenia (FDA, 2022a). The Food and Drug Administration (FDA) review of the safety database and of case narratives concluded that no embolic events were suggestive of TTS.

In the Phase 3 study COV3001, one Ad26.COV2.S recipient developed venous transverse sinus thrombosis and cerebral hemorrhage, which was confirmed as TTS (FDA, 2021b), meeting Brighton Collaboration criteria Level 1 (Chen and Buttery, 2021) and Centers for Disease Control and Prevention (CDC) Criteria Tier 1. In the Phase 3 study COV3009 (FDA, 2021b), which evaluated the efficacy and safety of a booster dose of Ad26.COV2.S, one recipient presented with DVT in combination with thrombocytopenia, Brighton Collaboration criteria Level 3. No Ad26.COV2.S recipients met the Brighton Collaboration criteria Level 1 or CDC Criteria Tier 1 for TTS (Chen and Buttery, 2021) in COV3001.

In the briefing document provided on NVX-CoV2373⁵ for FDA’s Vaccines and Related Biological Products Advisory Committee meeting, Novavax stated that no cases of TTS had been reported in the clinical trial evaluating its vaccine (FDA, 2022a).

Observational Studies

Table 5-1 presents three studies that contributed to the causality assessment.

Using electronic health record data from four hospitals in England and a self-controlled case series (SCCS) design, Higgins et al. (2022) evaluated the risk of TTS associated with the primary series of ChAdOx1-S and BNT162b2. TTS was defined as any acute thrombotic event associated with new onset of thrombocytopenia, defined as platelet count less than 150 × 10⁹ per L. The study population included 170 adults admitted to a hospital between January and March 2021. They found no increased risk of TTS in days 4–13, 14–27, 28–41, or 4–27 after the first dose of BNT162b2 for the overall population or subgroups defined by age (relative incidence [RI] 0.82, 95% confidence interval [CI]: 0.38–1.75) (Higgins et al., 2022). There was an increased risk of TTS on days 4–27 after the first dose of ChAdOx1-S for the subgroup of individuals aged 18–39, but this finding was not significant for the overall study population. Strengths of the study included the definition of the outcome, based on platelet counts, and the presentation of data stratified by age group. The findings were limited by the relatively small sample size available, which may have resulted in insufficient power to detect significant differences in TTS risk associated with ChAdOx1-S in the overall population.

___________________

³ Refers to the COVID-19 vaccine manufactured by Pfizer-BioNTech under the name Comirnaty^®.

⁴ Refers to the COVID-19 vaccine manufactured by Moderna under the name Spikevax^®.

⁵ Refers to the COVID-19 vaccine manufactured by Novavax.

TABLE 5-1 Epidemiological Studies in the Thrombosis with Thrombocytopenia Evidence Review

NOTES: BNT162b2 refers to the COVID-19 vaccine manufactured by Pfizer-BioNTech under the name Comirnaty^®. mRNA-1273 refers to the COVID-19 vaccine manufactured by Moderna under the name Spikevax^®. Number of events refers to events in vaccinees only. CI: confidence interval; RI: relative incidence; RR: risk ratio.

SOURCES: Andrews et al., 2022; Higgins et al., 2022; Klein et al., 2021.

Andrews et al. (2022) leveraged a national database of inpatient admissions in England to study the risk of TTS associated with the primary series of ChAdOx1-S and BNT162b2. TTS was defined as having a diagnosis code for thrombocytopenia and for a thrombotic event, including cerebral venous thrombosis, thrombophlebitis, deep venous thrombosis, splanchnic vein thrombosis, or PE (Andrews et al., 2022). Individuals with either diagnosis code in the year before the observation period were excluded to ensure that the study captured incident cases. Poisson regression models were used to compare the incidence of events after vaccination with that of unvaccinated individuals. The study included over 27 million individuals aged 15+ who were vaccinated between December 1, 2020, and April 18, 2021. Among those aged 15–39 and 65+, the first dose of BNT162b2 was not associated with an increased risk of TTS (Andrews et al., 2022). However, those aged 40–64 had an increased risk in days 14–27 after vaccination (RI 2.7, 95% CI: 1.1–7.1). This effect was not detected on days 0–13 or 4–13, due to insufficient number of events, or after the 28th day. There was a pronounced increase in the risk of TTS on days 4–13 and 14–37 after the first dose of ChAdOx1-S on individuals aged 15–39 and 40–64 but not for those aged 65+. The findings of the study are limited by the unavailability of platelet counts and the inclusion in the outcome definition of certain thrombotic events that are not typical manifestations of TTS, such as thrombophlebitis. Combined, these two limitations in the definition of the outcome may have resulted in an overestimation of the incidence of TTS.

Klein et al. (2021) leveraged data from the Vaccine Safety Datalink, which compiles data from eight U.S. health plans, to study the risk of 23 outcomes with BNT162b2 and mRNA-1273. TTS was defined as having an emergency room or inpatient diagnosis code for cerebral venous sinus thrombosis, splanchnic vein thrombosis, and arterial thrombosis and a platelet count less than 150 × 10⁹ per L (Klein et al., 2021). Poisson regression compared the risk of the outcome in the 21 days after the first or the second dose with that of comparators who were in days 22–42 after their most recent vaccination. Analyses were performed for the combination of the first and second doses of both BNT162b2 and mRNA-1273 combined. The study sample included 11,845,128 doses given to individuals aged 12+; everyone under 18 received BNT162b2. Individuals who had COVID-19 in the 30 days before vaccination were excluded from analyses. No significant differences were observed in the risk in days 1–21 post-vaccination compared to days 22–42 (Klein et al., 2021). The findings are limited by the combination of the risk period after the first and second doses and the lack of reporting of separate results for BNT162b2 and

mRNA-1273. The definition of the outcome was a strength of the study, as it leverages platelet counts and limits the list of thrombotic events to the common clinical manifestations of TTS.

Pharmacovigilance Data

Shortly after the reports on TTS with ChAdOx1-S became available, a report describing 12 U.S. patients with cerebral venous sinus thrombosis and thrombocytopenia 6–15 days after Ad26.COV2.S appeared (See et al., 2021). These patients had similar clinical manifestations as those with TTS after ChAdOx1-S and comparable platelet counts, elevated D-dimer levels, and positive testing for anti-PF4 antibodies. A follow-up study describing TTS cases reported to the U.S. Vaccine Adverse Event Reporting System (VAERS) between December 2020 and August 2021 was published in 2022. From 1,122 reports originally identified as potential TTS, 57 were determined to meet the case definition; 54 of them were after Ad26.COV2.S and three after mRNA vaccines (See et al., 2022). These case counts translated into reporting rates of 3.83 TTS cases per million doses of Ad26.COV2.S and 0.00855 TTS cases per million doses of mRNA vaccines. Reporting rates of TTS after Ad26.COV2.S were particularly pronounced among adults aged 18–49. No TTS events were reported for individuals under 18, although BNT162b2 was the only one authorized for use in the pediatric population during the study period (See et al., 2022).

Postmarketing reports of TTS after Ad26.COV2.S led to a pause in the use of the vaccine (April 13–23, 2021) and triggered an investigation by FDA and CDC (FDA, 2021c). In the most updated analyses, FDA reviewed TTS cases reported to VAERS through March 18, 2022, identifying 60 confirmed cases after Ad26.COV2.S, which translated into reporting rates of 3.23 cases per million doses (FDA, 2022a).

From Evidence to Conclusions

The three observational studies (Andrews et al., 2022; Klein et al., 2021; Li et al., 2022) considered in the causality assessment failed to find an association between mRNA vaccinations and TTS. An analysis of cases of TTS reported to VAERS found only three after mRNA vaccination (See et al., 2022), translating into a reporting rate of 0.00855 per million doses, which the committee interpreted as likely representative of the background rate in the general population.

The committee was not able to identify any data from comparative epidemiology studies on the association between Ad26.COV2.S and TTS. A study from VAERS estimated a reporting rate of 3.83 cases per 1 million doses of Ad26.COV2.S (See et al., 2022); these findings are consistent with the FDA evaluation of VAERS data (FDA, 2022b). The presence of anti-PF4 antibodies in individuals presenting with TTS after Ad26.COV2.S was deemed strong mechanistic evidence associating that vaccine with TTS, particularly when similar mechanistic data associating the ChAdOx1-S vaccine with TTS are taken into consideration. No evidence was available on a potential association with NVX-CoV2373.

IMMUNE THROMBOCYTOPENIC PURPURA

Background

ITP is an autoimmune disorder characterized primarily by a low platelet count, which can be associated with purpura and hemorrhagic episodes. It is often diagnosed through the exclusion of other causes of thrombocytopenia. IgG autoantibodies sensitize circulating platelets, leading to their accelerated removal by macrophages in the spleen and other components of the monocyte-macrophage system. Bone marrow responds by increasing platelet production. ITP is commonly observed in healthy children and young adults, often after a viral infection. The epidemiology varies; in children, spontaneous remission is common, whereas in adults, remission is rare, and patients are typically treated with a variety of therapies, including corticosteroids, rituximab, and thrombopoietin-mimetic agents.

The diagnosis of ITP is defined by a platelet count <100 × 10⁹ per L. Platelet counts of 100–150 × 10⁹ per L are frequently encountered in apparently healthy individuals, and most individuals in this range are unlikely to develop more severe thrombocytopenia (Rodeghiero et al., 2009). The disease can be classified into different phases based on the duration postdiagnosis: newly diagnosed (within the first 3 months), persistent (3–12 months), chronic (over 12 months), and refractory (failure of splenectomy). ITP occurs with infections, such as HIV; malignancies, such as lymphoma; and autoimmune diseases, such as systemic lupus erythematosus. Drug-induced ITP is also notable, with several medications implicated in it.

Epidemiologically, the acute form of ITP affects children and adults, but in children, it is relatively benign, often resolving spontaneously within 3 months. Chronic ITP more frequently affects adults primarily aged 20–50 years, with a higher prevalence in women, and may present with prolonged bleeding episodes and fluctuating platelet counts.

Mechanisms

ITP is highlighted by multiple immunological mechanisms, predominantly involving producing autoantibodies against platelet antigens. This autoimmune response is primarily driven by IgG autoantibodies targeting platelet membrane glycoproteins, such as GPIIb/IIIa and GPIb/IX (Kremer et al., 2022). Binding these autoantibodies to platelets tags them for destruction by the spleen’s macrophages, leading to a precipitous decline in platelet count. This process is intricately associated with the adaptive immune system, where B-lymphocytes play a pivotal role in autoantibody production, and T-lymphocytes may contribute to the loss of tolerance to platelet antigens (Audia et al., 2021).

A particularly intriguing aspect of ITP is its association with vaccines, known as “vaccine-induced ITP.” The exact immunological pathways are still being elucidated, but several hypotheses have been proposed. One suggests molecular mimicry, where vaccine antigens share structural similarities with platelet antigens, leading to cross-reactive immune responses (Segal and Shoenfeld, 2018). Another theory involves adjuvants, which can increase the immune response, potentially breaking the tolerance to self-antigens such as those found on platelets (McGonagle et al., 2021). Furthermore, the polyclonal activation of B cells by vaccines may inadvertently lead to producing autoantibodies against platelet antigens.

This ITP variant, although relatively rare, has been reported after various vaccinations, such as measles-mumps-rubella, varicella, and COVID-19 (Thomas et al., 2021). These cases are characterized by the rapid onset of thrombocytopenia, often within days to weeks. The immunological response to the vaccine’s vector may trigger an autoimmune reaction in predisposed individuals, leading to platelet destruction. However, the absolute risk remains low, and it is rarely associated with significant bleeding.

Recurrent severe thrombocytopenia has been observed in 6.1 to 17 percent of patients with pre-existing ITP following vaccination against SARS-CoV-2 (Kuter, 2021; Lee et al., 2022; Mori et al., 2023). Most of these patients quickly recover with rescue therapy, usually consisting of corticosteroids with or without intravenous immunoglobulin, and major bleeding is uncommon. It is generally recommended that patients with pre-existing ITP still get vaccinated but that platelet counts should be monitored afterward. Recurrent ITP is not considered further in the text below.

Epidemiological Evidence

Clinical trial results submitted to FDA for Emergency Use Authorization (EUA) and/or full approval do not indicate a signal regarding ITP and any of the vaccines under study (FDA, 2021d, 2023b,c,d). Table 5-2 presents seven studies that contributed to the causality assessment.

Shoaibi et al. (2023) followed an SCCS design to evaluate the risk of ITP after mRNA vaccines among Medicare beneficiaries aged 65+. Diagnosis codes from inpatient and outpatient claims were extracted; a medical review of health records of a random sample of cases was performed to validate the claims-based definition. From 91 cases of ITP identified in claims data and with health records available, only two were adjudicated as confirmed, one as probable, and six as possible. These statistics translated into a predictive positive value of 4 percent (1.37–11.11 percent), showing the high potential for misclassification based on diagnosis codes (Shoaibi et al., 2023). The limited validity of diagnosis codes in the detection of ITP was considered a major limitation of all studies that used them. As a result, the committee assigned limited weight to the epidemiology evidence in the causality conclusion. Shoaibi et al. (2023) found no association between the primary series of the BNT162b2 and mRNA-1273 and ITP; results were reported combining first and second doses under the primary series. No association was found between a booster dose of BNT162b2 and mRNA-1273 and the risk of ITP; however, the booster analysis was likely underpowered.

Two studies led by Simpson used National Health Service data from Scotland to evaluate the risk of ITP among other outcomes after administration of ChAdOx1-S and BNT162b2. The first study limited analyses to the first dose of each vaccine (Simpson et al., 2021), and the second evaluated risk after the second dose (Simpson et al., 2022). Both analyses constrained sampling to individuals aged 16+ and used read codes (equivalent to diagnosis codes) to define ITP. The assessment of the risk of ITP after first doses employed a matched case control nested within a cohort design; this analysis was complemented by a sensitivity analysis after an SCCS design (Simpson et al., 2021). The study evaluating risk of events after the second dose followed an SCCS design (Simpson et al., 2021). No significant association was found between the first dose of BNT162b2 and ITP (relative risk [RR] 0.54, 95% CI: 0.10–3.02) (Simpson et al., 2021). The risk of ITP did not significantly differ between days 0–27 after the second dose of BNT162b2 and the baseline period (Simpson et al., 2022). However, these findings are limited by a small number of ITP events (nine). Using Read codes is a major limitation of the analysis. The authors tried to overcome this limitation by the reporting of platelet counts for individuals with post-vaccination ITP; however, data were only available for a nonrepresentative share of cases.

TABLE 5-2 Epidemiological Studies in the Immune Thrombocytopenic Purpura Evidence Review

NOTES: BNT162b2 refers to the COVID-19 vaccine manufactured by Pfizer-BioNTech under the name Comirnaty^®. mRNA-1273 refers to the COVID-19 vaccine manufactured by Moderna under the name Spikevax^®. Shoaibi et al. (2023) combined the number of BNT162b2 and mRNA-1273 vaccinees. The primary series for mRNA vaccines is two doses. Number of events refers to events in vaccinees only. CI: confidence interval; IRR: incidence rate ratio; RR: relative risk; SIR: standardized incidence ratio.

SOURCES: Burn et al., 2022a,b; Klein et al., 2021; Shoaibi et al., 2023; Simpson et al., 2021, 2022; Torabi et al., 2022.

Two studies led by Burn compared the risk of ITP among other outcomes after ChAdOx1-S and BNT162b2 against historical rates. The first used UK electronic health record data for individuals aged 20+ (Burn et al., 2022a). ITP events were defined using diagnosis codes; however, the specific codes used were not reported. The risk of ITP in the 28 days after either the first or the second dose of ChAdOx1-S and BNT162b2 was compared against individuals in the database from January 2017 to December 2019. The committee noted that the use of historical background rates could have biased results, as the background incidence of ITP may have been lower during the COVID-19 pandemic compared to the pre-pandemic period (Sakurai et al., 2023). Individuals with a recorded diagnosis of ITP in the year before vaccination were excluded. ITP events after the first dose of ChAdOx1-S significantly exceeded the expectation based on historical rates (standardized incidence ratio [SIR] 1.79, 95% CI: 1.33–2.39) (Burn et al., 2022a). The number of ITP events after the second dose of ChAdOx1-S or the first or second dose of BNT162b2 was not significantly different from expectations. These results were limited by the lack of adjustment for potential differences in clinical characteristics between the historical comparators and the vaccinated people and the low validity of using diagnosis codes.

The second study by Burn et al. (2022b) used electronic health record data from Spain to compare the cases of ITP observed in the 21 days after the first dose of ChAdOx1-S or the first and second doses of BNT162b2 against historical rates from 2017. No definition was provided for how ITP was determined. Analyses were limited to individuals aged 20+, and those with ITP in the year before vaccination were excluded. No adjustment for potential differences in clinical characteristics between vaccinated subjects and historical comparators was conducted. ITP cases after the first dose of ChAdOx1-S (n = 12) were significantly lower than expected based on historical rates (SIR 0.48, 95% CI: 0.27–0.85) (Burn et al., 2022b). ITP cases after the first dose of BNT162b2 (n = 97) were not significantly different from expectations; however, those after the second dose of BNT162b2 (n = 61) were significantly lower than expected (SIR 0.69, 95% CI: 0.53–0.88) (Burn et al., 2022b). The association of the first dose of ChAdOx1-S and the second dose of BNT162b2 with a decreased risk of ITP may indicate residual confounding.

Klein et al. (2021) described under the TTS section, compared the risk of ITP after first and second doses of BNT162b2 and mRNA-1273 combined. ITP was defined using diagnosis codes from emergency department and inpatient and outpatient claims (Klein et al., 2021). No significant differences were observed in the risk of ITP in days 1–21 post-vaccination compared to days 22–42 (RR 1.12, 95% CI: 0.65–1.97).

Torabi et al. (2022) used electronic health record data from Wales and evaluated the risk of ITP after first and second doses of BNT162b2. ITP was defined using diagnosis codes. The study incorporated an SCCS method and compared the risk of events in the 28 days after vaccination against a 90-day prevaccination baseline period and a post-vaccination control period (median of 72 days). The study population included 2.1 million individuals aged 16+. They found an increased risk of ITP on days 0–7 after the first dose of BNT162b2 (incidence rate ratio [IRR] 2.80, 95% CI 1.21–6.49) but not on days 8–14, 15–21, or 22–28, nor was there an increased risk 0–7 days after dose 2 or a booster dose (Torabi et al., 2022). Their findings are limited by the number of events, which was fewer than 10 for each of the 7-day intervals (Torabi et al., 2022), and the high misclassification associated with diagnosis codes.

Pharmacovigilance and Surveillance

In the briefing documents provided by the sponsor for the advisory committee convened to review the EUA amendment for the booster dose of Ad26.COV2.S, the manufacturer disclosed that ITP had been reported in the post-marketing setting (FDA, 2021b). The manufacturer reported analyzing a U.S.-based claims database and found an increased risk of ITP within 28, 42, and 90 days of vaccination, using both SCCS and comparative designs (RR estimates 1.86–2.22) (FDA, 2021b). This study was only briefly described by the manufacturer in the EUA addendum and not published; as a result, the committee was not able to evaluate the definition of ITP used and methodology employed. The manufacturer also conducted an analysis of ITP reports submitted to VAERS by July 31, 2021, and estimated an observed-to-expected ratio of 3.6 (95% CI: 3.0–4.1) for individuals 18–59 and 3.0 (95% CI: 2.4–3.8) for individuals 60+. In this evaluation, the manufacturer used a case definition of ITP that included reports with platelet counts below 100 × 10⁹ per L (FDA, 2021b).

FDA conducted an independent investigation of ITP reports submitted to VAERS by September 30, 2021, and estimated an overall observed-to-expected ratio of 4.04 (95% CI: 3.42–4.72) in the 28 days after Ad26.COV2.S

(FDA, 2022a). This investigation, however, defined ITP as having a platelet count less than 150 × 10⁹ per L or a diagnosis of thrombocytopenia without a documented platelet count. The committee considered this definition too imprecise, however, given the criteria for the diagnosis of ITP noted at the beginning of this section. An update using reports submitted to VAERS through December 2021 was published in the peer-reviewed literature (Woo and Dimova, 2022). In addition to the earlier definition of ITP, which was considered the base case, this FDA study also included a “narrow” definition, which only included cases with documented platelet counts below 100 × 10⁹ per L; the observed-to-expected ratio decreased to 1.55 (95% CI: 1.20–1.98). The authors acknowledged multiple limitations, however, including the lack of adjudication of cases by hematologists. Several reports mentioned a “history of thrombocytopenia,” and five had experienced “clinically significant thrombotic or thromboembolic events” without meeting the case definition of TTS, an atypical presentation for ITP. Given these methodologic concerns, the committee felt that the pharmacovigilance data were insufficient to support a causal relationship between Ad26.COV2.S and ITP.

FDA also reviewed pharmacovigilance data submitted to VAERS through February 4, 2021, looking at thrombocytopenia and ITP after mRNA vaccines (Welsh et al., 2021) and did not find an increased rate of reported cases.

From Evidence to Conclusions

With the exception of Shoaibi et al. (2023), who performed a medical review of electronic health record data, the totality of the epidemiology evidence on the potential association of BNT162b2 and mRNA-1273 and ITP is based on diagnosis codes. Shoaibi et al. (2023) estimated that the use of diagnosis codes had a positive predictive value of 4.0 percent for case identification. As a result, the committee deemed the use of diagnosis codes in the ITP case definition a major limitation of the studies.

No published comparative epidemiology assessments evaluated the potential association of Ad26.COV2.S with ITP. A manufacturer analysis (FDA, 2021b) of a U.S.-based claims database found an increased risk. As this study was not published, the committee was unable to assess the definition of ITP and methodological rigor. The manufacturer analysis of VAERS data through July 2021 and the initial FDA evaluation of reports through September were considered superseded by the FDA evaluation that included reports through December 2021. The narrow definition of ITP (platelet counts below 100 × 10⁹ per L) yielded an observed-to-expected ratio of 1.55, considerably lower than that estimated with the definition that also included reports with platelet counts of 100–150 × 10⁹ per L. Additionally, there was a lack of evidence on a potential mechanism of action linking Ad26.COV2.S with ITP. No evidence was available on the association of NVX-CoV2373 with ITP.

CAPILLARY LEAK SYNDROME

Background

CLS, also known as “Clarkson disease,” is a complex and potentially lethal condition characterized by an initial phase of nonspecific symptoms followed by the hallmark features of diffuse severe edema and hypovolemia, hemoconcentration, and hypoalbuminemia (Bichon et al., 2021). This condition is often triggered by factors such as drugs (including antitumor therapies), malignancy, infections (predominantly viral), and inflammatory diseases. Its pathophysiology involves severe, transient, and multifactorial endothelial disruption, the mechanisms of which remain unclear. Treatment is primarily empirical and symptomatic during the acute phase, with the addition of drugs that amplify cyclic adenosine monophosphate levels in severe cases. Prophylactic monthly polyvalent immunoglobulins are used to prevent relapses (Bichon et al., 2021; Siddall et al., 2017).

U.S. reports exist of fatal exacerbations of CLS in patients with mild-to-moderate COVID-19 symptoms (Bichon et al., 2021). In these cases, the clinical diagnostic triad for CLS (hypotension, hemoconcentration, hypoalbuminemia) was observed, indicating that individuals with known or suspected CLS may be at increased risk of a disease flare in the context of COVID-19 (Felten et al., 2021). The cytokine storm associated with COVID-19 is thought to potentially lead to a CLS flare; alternatively, the virus may directly affect endothelial cells (Mohseni Afshar et al., 2023). This highlights the need for increased vigilance in patients with CLS or related inflammatory diseases during the COVID-19 pandemic.

Diagnostically, CLS is identified clinically based on a symptomatic triad of hypotension, hemoconcentration, and hypoalbuminemia resulting from fluid extravasation. Blood tests are important for diagnosing CLS, looking for increased levels of hematocrit and hemoglobin, and low blood protein levels. The presence of abnormal monoclonal gammopathy (M protein) is also a diagnostic consideration (Kapoor et al., 2010).

Systemic CLS is a rare disorder, affecting fewer than 500 people worldwide. It predominantly occurs in middle-aged adults and is very rare in children (NORD, 2020). However, the actual incidence may be higher due to potential misdiagnosis (Kapoor et al., 2010).

Mechanisms

CLS involves an increase in capillary permeability to proteins, leading to the loss of protein-rich fluid from the intravascular space to the interstitial space (Siddall et al., 2017). This phenomenon is most commonly associated with sepsis but can occur in a variety of other conditions, such as idiopathic systemic CLS, engraftment syndrome, differentiation syndrome, ovarian hyperstimulation syndrome, hemophagocytic lymph histiocytosis,

viral hemorrhagic fevers, autoimmune diseases, snakebite, ricin poisoning, and adverse effects from certain drugs, such as some interleukins, monoclonal antibodies, and gemcitabine (Siddall et al., 2017). The diseases associated with CLS, including sepsis, often manifest with diffuse pitting edema, exudative serous cavity effusions, noncardiogenic pulmonary edema, hypotension, and sometimes hypovolemic shock with multiple-organ failure. Acute kidney injury is a common complication in these conditions, and cytokines are believed to play a significant role in acute kidney injury in CLS. Fluid management is critical in treating CLS, as both hypovolemia and hypotension can cause organ injury, and capillary leakage of administered fluid can worsen organ edema, leading to progressive organ injury (Ruggiero et al., 2022).

CLS is also strongly associated with cytokine activity states and an underrecognized early immune effect of checkpoint inhibitor treatment, which is more typically associated with cellular immune responses (Ruggiero et al., 2022). The interaction between checkpoint inhibitors, cellular immunity, cytokine action, and endothelial damage has been noted in individuals with CLS after checkpoint inhibitor treatment. This suggests that CLS may be an unusual effect of immunotherapy, resulting from complex interactions between cellular immunity and cytokine activation, and its expression likely depends on inherent immune variation (Wong So et al., 2023).

Epidemiological Evidence

Clinical trial results submitted to FDA for EUA and/or full approval do not indicate a signal regarding CLS and any of the vaccines under study (FDA, 2021d, 2023b,c,d). The committee failed to identify any comparative epidemiology studies evaluating CLS after COVID-19 vaccination. A study from the European pharmacovigilance database EudraVigilance reported that CLS emerged as a new adverse event after immunization associated with COVID-19 vaccination. Between January 1, 2021, and January 14, 2022, there were 36 CLS case reports associated with BNT162b2, three with mRNA-1273, 36 with ChAdOx1-S, and nine with Ad26.COV2.S. A disproportionality analysis of these reports associated mRNA vaccines with a decreased CLS reporting probability compared to viral vaccines (rate of return 0.5, 95% CI: 0.3–0.7) (Ruggiero et al., 2022). This study evaluated the onset of CLS after COVID-19 mRNA vaccines compared to viral vector vaccines. Cytokine release after T cell activation could be involved in CLS, but a precise mechanism has not yet been identified (Ruggiero et al., 2022).

From Evidence to Conclusions

No comparative epidemiology studies evaluated the risk of CLS with COVID-19 vaccinations. Pharmacovigilance data available were inconclusive. Despite plausible mechanistic hypotheses that link potential mechanisms for CLS with the COVID-19 vaccinations (e.g., cytokine activation, endothelial cell perturbation), no available mechanistic data clearly link vaccination with the clinical development of CLS.

REFERENCES

Andrews, N. J., J. Stowe, M. E. Ramsay, and E. Miller. 2022. Risk of venous thrombotic events and thrombocytopenia in sequential time periods after ChAdOx1 and BNT162b2 COVID-19 vaccines: A national cohort study in England. Lancet Regional Health—Europe 13:100260. https://doi.org/10.1016/j.lanepe.2021.100260.

Audia, S., M. Mahevas, M. Nivet, S. Ouandji, M. Ciudad, and B. Bonnotte. 2021. Immune thrombocytopenia: Recent advances in pathogenesis and treatments. HemaSphere 5(6):e574. https://doi.org/10.1097/HS9.0000000000000574.

Bichon, A., J. Bourenne, M. Gainnier, and J. Carvelli. 2021. Capillary leak syndrome: State of the art in 2021. Journal of Internal Medicine 42(11):789–796. https://doi.org/10.1016/j.revmed.2021.05.012.

Burn, E., X. Li, A. Delmestri, N. Jones, T. Duarte-Salles, C. Reyes, E. Martinez-Hernandez, E. Marti, K. M. C. Verhamme, P. R. Rijnbeek, V. Y. Strauss, and D. Prieto-Alhambra. 2022a. Thrombosis and thrombocytopenia after vaccination against and infection with SARS-CoV-2 in the United Kingdom. Nature Communications 13(1):7167. https://doi.org/10.1038/s41467-022-34668-w.

Burn, E., E. Roel, A. Pistillo, S. Fernández-Bertolín, M. Aragón, B. Raventós, C. Reyes, K. Verhamme, P. Rijnbeek, X. Li, V. Y. Strauss, D. Prieto-Alhambra, and T. Duarte-Salles. 2022b. Thrombosis and thrombocytopenia after vaccination against and infection with SARS-CoV-2 in Catalonia, Spain. Nature Communications 13(1):7169. https://doi.org/10.1038/s41467-022-34669-9.

Chen, R., and J. Buttery. 2021. Updated Brighton Collaboration case definition of thrombosis with thrombocytopenia syndrome (TTS). https://brightoncollaboration.org/wp-content/uploads/2023/08/TTS-Updated-Brighton-Collaboration-Case-Defintion-Draft-Nov-11-2021.pdf (accessed December 12, 2023).

FDA (Food and Drug Administration). 2021a. BLA clinical review memorandum—COMIRNATY. https://www.fda.gov/media/152256/download (accessed December 5, 2023).

FDA. 2021b. Vaccines and Related Biological Products Advisory Committee October 14–15, 2021 meeting announcement. https://www.fda.gov/advisory-committees/advisory-committee-calendar/vaccines-and-related-biological-products-advisory-committee-october-14-15-2021-meeting-announcement (accessed December 12, 2023).

FDA. 2021c. FDA and CDC lift recommended pause on Johnson & Johnson (Janssen) COVID-19 vaccine use following thorough safety review. https://www.fda.gov/news-events/press-announcements/fda-and-cdc-lift-recommended-pause-johnson-johnson-janssen-covid-19-vaccine-use-following-thorough (accessed March 8, 2024).

FDA. 2021d. Emergency use authorization (EUA) amendment for an unapproved product review memorandum. https://www.fda.gov/media/153439/download (accessed May 3, 2023).

FDA. 2022a. CBER assessment of new safety information on immune thrombocytopenia (ITP) following administration of the Janssen COVID-19 vaccine. Center for Biologics Evaluation and Research.

FDA. 2022b. BLA clinical review memorandum—SPIKEVAX. https://www.fda.gov/media/156342/download (accessed December 5, 2023).

FDA. 2023a. Coronavirus (COVID-19) update: FDA authorizes changes to simplify use of bivalent mRNA COVID-19 vaccines. https://www.fda.gov/news-events/press-announcements/coronavirus-covid-19-update-fda-authorizes-changes-simplify-use-bivalent-mrna-covid-19-vaccines (accessed December 12, 2023).

FDA. 2023b. BLA clinical review memorandum—COMIRNATY. https://www.fda.gov/media/172333/download?attachment (accessed December 5, 2023).

FDA. 2023c. BLA clinical review memorandum—SPIKEVAX. https://www.fda.gov/media/172357/download?attachment (accessed December 5, 2023).

FDA. 2023d. Emergency use authorization (EUA) for an unapproved product review memorandum. https://www.fda.gov/media/168233/download?attachment (accessed December 5, 2023).

Felten, R., M. Scherlinger, A. Guffroy, V. Poindron, A. Meyer, M. Giannini, A. S. Korganow, C. Sordet, E. Chatelus, R. M. Javier, A. Meyer, L. Pijnenburg, J. F. Kleinmann, J. E. Gottenberg, J. Sibilia, T. Martin, and L. Arnaud. 2021. Incidence and predictors of COVID-19 and flares in patients with rare autoimmune diseases: A systematic survey and serological study at a national reference center in France. Arthritis Research & Therapy 23(1):188. https://doi.org/10.1186/s13075-021-02565-0.

Greinacher, A., T. Thiele, T. E. Warkentin, K. Weisser, P. A. Kyrle, and S. Eichinger. 2021. Thrombotic thrombocytopenia after ChAdOx1 nCoV-19 vaccination. New England Journal of Medicine 384(22):2092–2101. https://doi.org/10.1056/NEJMoa2104840.

Higgins, H., N. Andrews, J. Stowe, G. Amirthalingam, M. Ramsay, G. Bahra, A. Hackett, K. A. Breen, M. Desborough, D. Khan, H. Leary, C. Sweeney, E. Hutchinson, S. E. Shapiro, C. Lees, J. Dhanapal, P. K. MacCallum, S. Burke, V. McDonald, N. M. A. Entwistle, S. Booth, C. J. Atchison, and B. J. Hunt. 2022. Risk of thrombosis with thrombocytopenia syndrome after COVID-19 vaccination before the recognition of vaccine-induced thrombocytopenia and thrombosis: A self-controlled case series study in England. Research and Practice in Thrombosis and Haemostasis 6(3):e12698. https://doi.org/10.1002/rth2.12698.

Huynh, A., D. M. Arnold, J. G. Kelton, J. W. Smith, P. Horsewood, R. Clare, A. Guarné, and I. Nazy. 2019. Characterization of platelet factor 4 amino acids that bind pathogenic antibodies in heparin-induced thrombocytopenia. Journal of Thrombosis and Haemostasis 17(2):389–399. https://doi.org/https://doi.org/10.1111/jth.14369.

Huynh, A., D. M. Arnold, N. Ivetic, R. Clare, M. Hadzi-Tosev, Y. Liu, J. W. Smith, A. L. Bissola, M. Daka, J. G. Kelton, and I. Nazy. 2023. Antibodies against platelet factor 4 and the risk of cerebral venous sinus thrombosis in patients with vaccine-induced immune thrombotic thrombocytopenia. Journal of Thrombosis and Haemostasis 21(10):2833–2843. https://doi.org/10.1016/j.jtha.2023.06.026.

Jay, R. M., and T. E. Warkentin. 2008. Fatal heparin-induced thrombocytopenia (HIT) during warfarin thromboprophylaxis after orthopedic surgery: Another example of “spontaneous” HIT? Journal of Thrombosis and Haemostasis 6(9):1598–1600. https://doi.org/10.1111/j.1538-7836.2008.03040.x.

Kanack, A. J., and A. Padmanabhan. 2022. Vaccine-induced immune thrombotic thrombocytopenia. Best Practice & Research: Clinical Haematology 35(3):101381. https://doi.org/10.1016/j.beha.2022.101381.

Kapoor, P., P. T. Greipp, E. W. Schaefer, S. J. Mandrekar, A. H. Kamal, N. C. Gonzalez-Paz, S. Kumar, and P. R. Greipp. 2010. Idiopathic systemic capillary leak syndrome (Clarkson’s Disease): The Mayo Clinic experience. Mayo Clinic Proceedings 85(10):905–912. https://doi.org/10.4065/mcp.2010.0159.

Klein, N. P., N. Lewis, K. Goddard, B. Fireman, O. Zerbo, K. E. Hanson, J. G. Donahue, E. O. Kharbanda, A. Naleway, J. C. Nelson, S. Xu, W. K. Yih, J. M. Glanz, J. T. B. Williams, S. J. Hambidge, B. J. Lewin, T. T. Shimabukuro, F. DeStefano, and E. S. Weintraub. 2021. Surveillance for adverse events after COVID-19 mRNA vaccination. JAMA 326(14):1390–1399. https://doi.org/10.1001/jama.2021.15072.

Kremer, V., L. de Chaisemartin, and F. Jonsson. 2022. The role of neutrophils in antibody-driven autoimmune cytopenias. International Journal of Biochemistry and Cell Biology 147:106231. https://doi.org/10.1016/j.biocel.2022.106231.

Kuter, D. J. 2021. Exacerbation of immune thrombocytopenia following COVID-19 vaccination. British Journal of Haematology 195(3):365–370. https://doi.org/10.1111/bjh.17645.

Lee, E. J., M. Beltrami-Moreira, H. Al-Samkari, A. Cuker, J. DiRaimo, T. Gernsheimer, A. Kruse, C. Kessler, C. Kruse, A. D. Leavitt, A. I. Lee, H. A. Liebman, A. C. Newland, A. E. Ray, M. D. Tarantino, J. Thachil, D. J. Kuter, D. B. Cines, and J. B. Bussel. 2022. SARS-CoV-2 vaccination and ITP in patients with de novo or preexisting ITP. Blood 139(10):1564–1574. https://doi.org/10.1182/blood.2021013411.

Leung, H. H. L., J. Perdomo, Z. Ahmadi, S. S. Zheng, F. N. Rashid, A. Enjeti, S. B. Ting, J. J. H. Chong, and B. H. Chong. 2022. Netosis and thrombosis in vaccine-induced immune thrombotic thrombocytopenia. Nature Communications 13(1):5206. https://doi.org/10.1038/s41467-022-32946-1.

Li, X., E. Burn, T. Duarte-Salles, C. Yin, C. Reich, A. Delmestri, K. Verhamme, P. Rijnbeek, M. A. Suchard, K. Li, M. Mosseveld, L. H. John, M. A. Mayer, J. M. Ramirez-Anguita, C. Cohet, V. Strauss, and D. Prieto-Alhambra. 2022. Comparative risk of thrombosis with thrombocytopenia syndrome or thromboembolic events associated with different COVID-19 vaccines: International network cohort study from five European countries and the U.S. British Journal of Medicine 379:e071594. https://doi.org/10.1136/bmj-2022-071594.

McGonagle, D., G. De Marco, and C. Bridgewood. 2021. Mechanisms of immunothrombosis in vaccine-induced thrombotic thrombocytopenia (VITT) compared to natural SARS-CoV-2 infection. Journal of Autoimmunity 121:102662. https://doi.org/10.1016/j.jaut.2021.102662.

Mohseni Afshar, Z., M. Barary, A. Babazadeh, A. Tavakoli Pirzaman, R. Hosseinzadeh, A. Alijanpour, A. Allahgholipour, S. R. Miri, T. T. Sio, M. J. M. Sullman, K. Carson-Chahhoud, and S. Ebrahimpour. 2023. The role of cytokines and their antagonists in the treatment of COVID-19 patients. Reviews in Medical Virology 33(1):e2372. https://doi.org/10.1002/rmv.2372.

Mori, A., M. Onozawa, M. Kobayashi, S. Tsukamoto, H. Senjo, T. Ishio, E. Yokoyama, K. Izumiyama, M. Saito, H. Muraki, M. Morioka, T. Teshima, and T. Kondo. 2023. Humoral response to mRNA-based COVID-19 vaccine in patients with immune thrombocytopenia. British Journal of Haematology 200(6):717–721. https://doi.org/10.1111/bjh.18578.

NORD (National Organization for Rare Disorders). 2020. Systemic capillary leak syndrome. https://rarediseases.org/rarediseases/systemic-capillary-leak-syndrome (accessed February 22, 2024).

Pavord, S., M. Scully, B. J. Hunt, W. Lester, C. Bagot, B. Craven, A. Rampotas, G. Ambler, and M. Makris. 2021. Clinical features of vaccine-induced immune thrombocytopenia and thrombosis. New England Journal of Medicine 385(18):1680–1689. https://doi.org/10.1056/NEJMoa2109908.

Rodeghiero, F., R. Stasi, T. Gernsheimer, M. Michel, D. Provan, D. M. Arnold, J. B. Bussel, D. B. Cines, B. H. Chong, N. Cooper, B. Godeau, K. Lechner, M. G. Mazzucconi, R. McMillan, M. A. Sanz, P. Imbach, V. Blanchette, T. Kuhne, M. Ruggeri, and J. N. George. 2009. Standardization of terminology, definitions and outcome criteria in immune thrombocytopenic purpura of adults and children: Report from an international working group. Blood 113(11):2386–2393. https://doi.org/10.1182/blood-2008-07-162503.

Ruggiero, R., N. Balzano, R. Di Napoli, A. Mascolo, P. M. Berrino, C. Rafaniello, L. Sportiello, F. Rossi, and A. Capuano. 2022. Capillary leak syndrome following COVID-19 vaccination: Data from the European pharmacovigilance database EudraVigilance. Frontiers in Immunology 13:956825. https://doi.org/10.3389/fimmu.2022.956825.

Sakurai, M., Y. Kogure, K. Mizuno, E. Matsuki, and K. Kataoka. 2023. Long-term reduction in the incidence of aplastic anemia and immune thrombocytopenia during the COVID-19 pandemic. Haematologica 108(9):2546–2550. https://doi.org/10.3324/haematol.2022.282351.

Schultz, N. H., I. H. Sørvoll, A. E. Michelsen, L. A. Munthe, F. Lund-Johansen, M. T. Ahlen, M. Wiedmann, A.-H. Aamodt, T. H. Skattør, and G. E. Tjønnfjord. 2021. Thrombosis and thrombocytopenia after ChAdOx1 NCOV-19 vaccination. New England Journal of Medicine 384(22):2124–2130.

Scully, M., D. Singh, R. Lown, A. Poles, T. Solomon, M. Levi, D. Goldblatt, P. Kotoucek, W. Thomas, and W. Lester. 2021. Pathologic antibodies to platelet factor 4 after ChAdOx1 NCOV-19 vaccination. New England Journal of Medicine 384(23):2202–2211. https://doi.org/10.1056/NEJMoa2105385.

See, I., J. R. Su, A. Lale, E. J. Woo, A. Y. Guh, T. T. Shimabukuro, M. B. Streiff, A. K. Rao, A. P. Wheeler, S. F. Beavers, A. P. Durbin, K. Edwards, E. Miller, T. A. Harrington, A. Mba-Jonas, N. Nair, D. T. Nguyen, K. R. Talaat, V. C. Urrutia, S. C. Walker, C. B. Creech, T. A. Clark, F. DeStefano, and K. R. Broder. 2021. U.S. case reports of cerebral venous sinus thrombosis with thrombocytopenia after Ad26.COV2.S vaccination, March 2 to April 21, 2021. JAMA 325(24):2448–2456. https://doi.org/10.1001/jama.2021.7517.

See, I., A. Lale, P. Marquez, M. B. Streiff, A. P. Wheeler, N. K. Tepper, E. J. Woo, K. R. Broder, K. M. Edwards, R. Gallego, A. I. Geller, K. A. Jackson, S. Sharma, K. R. Talaat, E. B. Walter, I. J. Akpan, T. L. Ortel, V. C. Urrutia, S. C. Walker, J. C. Yui, T. T. Shimabukuro, A. Mba-Jonas, J. R. Su, and D. K. Shay. 2022. Case series of thrombosis with thrombocytopenia syndrome after COVID-19 vaccination—United States, December 2020 to August 2021. Annals of Internal Medicine 175(4):513–522. https://doi.org/10.7326/m21-4502.

Segal, Y., and Y. Shoenfeld. 2018. Vaccine-induced autoimmunity: The role of molecular mimicry and immune crossreaction. Cellular & Molecular Immunology 15(6):586–594. https://doi.org/10.1038/cmi.2017.151.

Shoaibi, A., P. C. Lloyd, H. L. Wong, T. C. Clarke, Y. Chillarige, R. Do, M. Hu, Y. Jiao, A. Kwist, A. Lindaas, K. Matuska, R. McEvoy, M. Ondari, S. Parulekar, X. Shi, J. Wang, Y. Lu, J. Obidi, C. K. Zhou, J. A. Kelman, R. A. Forshee, and S. A. Anderson. 2023. Evaluation of potential adverse events following COVID-19 mRNA vaccination among adults aged 65 years and older: Two self-controlled studies in the U.S. Vaccine 41(32):4666–4678. https://doi.org/10.1016/j.vaccine.2023.06.014.

Siddall, E., M. Khatri, and J. Radhakrishnan. 2017. Capillary leak syndrome: Etiologies, pathophysiology, and management. Kidney International 92(1):37–46. https://doi.org/10.1016/j.kint.2016.11.029.

Simpson, C. R., T. Shi, E. Vasileiou, S. V. Katikireddi, S. Kerr, E. Moore, C. McCowan, U. Agrawal, S. A. Shah, L. D. Ritchie, J. Murray, J. Pan, D. T. Bradley, S. J. Stock, R. Wood, A. Chuter, J. Beggs, H. R. Stagg, M. Joy, R. S. M. Tsang, S. de Lusignan, R. Hobbs, R. A. Lyons, F. Torabi, S. Bedston, M. O’Leary, A. Akbari, J. McMenamin, C. Robertson, and A. Sheikh. 2021. First-dose ChAdOx1 and BNT162b2 COVID-19 vaccines and thrombocytopenic, thromboembolic and hemorrhagic events in Scotland. Nature Medicine 27(7):1290–1297. https://doi.org/10.1038/s41591-021-01408-4.

Simpson, C. R., S. Kerr, S. V. Katikireddi, C. McCowan, L. D. Ritchie, J. Pan, S. J. Stock, I. Rudan, R. S. M. Tsang, S. de Lusignan, F. D. R. Hobbs, A. Akbari, R. A. Lyons, C. Robertson, and A. Sheikh. 2022. Second-dose ChAdOx1 and BNT162b2 COVID-19 vaccines and thrombocytopenic, thromboembolic and hemorrhagic events in Scotland. Nature Communications 13(1):4800. https://doi.org/10.1038/s41467-022-32264-6.

Singh, A., F. Toma, G. Uzun, T. R. Wagner, L. Pelzl, J. Zlamal, V. Freytag, K. Weich, S. Nowak-Harnau, U. Rothbauer, K. Althaus, and T. Bakchoul. 2022. The interaction between anti-PF4 antibodies and anticoagulants in vaccine-induced thrombotic thrombocytopenia. Blood 139(23):3430–3438. https://doi.org/10.1182/blood.2021013839.

Thomas, W., A. Albano, D. Kirkel, N. Rouhizad, and F. Arinze. 2021. Immune thrombocytopenic purpura following administration of mRNA-based SARS-CoV-2 and MMR vaccinations: A cautionary tale. Case Reports in Infectious Diseases 2021:2704249. https://doi.org/10.1155/2021/2704249.

Torabi, F., S. Bedston, E. Lowthian, A. Akbari, R. K. Owen, D. T. Bradley, U. Agrawal, P. Collins, R. Fry, L. J. Griffiths, J. Beggs, G. Davies, J. Hollinghurst, J. Lyons, H. Abbasizanjani, S. Cottrell, M. Perry, R. Roberts, A. Azcoaga-Lorenzo, A. F. Fagbamigbe, T. Shi, R. S. M. Tsang, C. Robertson, F. D. R. Hobbs, S. de Lusignan, C. McCowan, M. Gravenor, C. R. Simpson, A. Sheikh, and R. A. Lyons. 2022. Risk of thrombocytopenic, haemorrhagic and thromboembolic disorders following COVID-19 vaccination and positive test: A self-controlled case series analysis in Wales. Scientific Reports 12(1):16406. https://doi.org/10.1038/s41598-022-20118-6.

Warkentin, T. E. 2022. Platelet-activating anti-PF4 disorders: An overview. Seminars in Hematology 59(2):59–71. https://doi.org/10.1053/j.seminhematol.2022.02.005.

Warkentin, T. E., A. Greinacher, A. Koster, and A. M. Lincoff. 2008. Treatment and prevention of heparin-induced thrombocytopenia: American College of Chest Physicians evidence-based clinical practice guidelines (8th edition). Chest 133(6 Suppl):340s–380s. https://doi.org/10.1378/chest.08-0677.

Welsh, K. J., J. Baumblatt, W. Chege, R. Goud, and N. Nair. 2021. Thrombocytopenia including immune thrombocytopenia after receipt of mRNA COVID-19 vaccines reported to the Vaccine Adverse Event Reporting System (VAERS). Vaccine 39(25):3329–3332. https://doi.org/10.1016/j.vaccine.2021.04.054.

Wong So, J., F. Bouibede, A. P. Jonville-Bera, F. Maillot, F. Barbier, and B. Largeau. 2023. Immune checkpoint inhibitor-associated capillary leak syndrome: A systematic review and a worldwide pharmacovigilance study. Journal of Internal Medicine 294(1):58–68. https://doi.org/10.1111/joim.13641.

Woo, E. J., and R. B. Dimova. 2022. Thrombocytopenia after Ad.26.COV2.S COVID-19 vaccine: Reports to the Vaccine Adverse Event Reporting System. Vaccine 40(31):4116–4120. https://doi.org/10.1016/j.vaccine.2022.05.078.