High-Mobility Group Box 1 Protein (HMGB1) as a Novel Diagnostic Tool for Toxic Epidermal Necrolysis and Stevens-Johnson Syndrome

Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are rare but potentially life-threatening adverse drug reactions characterized by epidermal necrosis.¹ Skin manifestations of SJS and TEN at the early stage can be similar to those of other ordinal drug eruptions like erythema exudativum multiforme (EM) and morbilliform drug eruptions, which sometimes makes early diagnosis difficult.

Recent studies have shown that severely damaged cells, such as necrotic cells and apoptotic cells, release high-mobility group box 1 protein (HMGB1), a nonhistone nuclear protein with a dual function. Inside the cells, HMGB1 is a nuclear constituent loosely bound to chromatin, and it plays a role in transcriptional regulation, while outside the cells, HMGB1 serves as an activator of the inflammatory cascade.^2,3

To evaluate a link between HMGB1 and SJS and TEN, we measured HMGB1 levels in serum samples from 22 healthy control subjects (mean [SD] age, 31.2 [4.6] years; 9 men, 13 women), 11 patients with maculopapular drug eruptions (MPEs), 13 patients with EM, and 13 patients with SJS and/or TEN (Table). All diagnoses were based on the classification system proposed previously.⁴ Disease onset (day 1) in patients with SJS and/or TEN was defined as the day when mucocutaneous and/or ocular lesions occurred. (Of note, we collected the serum samples of SJS and/or TEN before the onset of SJS and/or TEN, when the patients might have been diagnosed as having EM or MPE at the time of the assay, but the patients then developed SJS and/or TEN.)

The serum HMGB1 levels were measured with an HMGB1 enzyme-linked immunosorbent assay kit (Shino-Test Co, Tokyo, Japan) according to the manufacturer's protocol. Informed consents were obtained from each patient, and the institutional review board or ethics committee of each institution approved the study protocol.

Serum HMGB1 levels of all healthy control subjects and patients with MPE and EM were lower than 6 ng/mL. In contrast, serum HMGB1 levels in patients with SJS and/or TEN at days −7 to −1, days 1 to 7, and days 8 to 21 after onset were significantly higher than those of both healthy controls and patients with MPE and EM (Figure 1). The sensitivity of the assay for SJS and/or TEN above the threshold level was 45.4%. Consistent with these results, immunohistochemical analysis for HMGB1 with anti-HMGB1 antibody levels (Upstate Biotechnology, Lake Placid, New York) and diaminobenzidine staining demonstrated strong positivity for HMGB1 at the nuclei and perinuclei of keratinocytes in the necrotic keratinocytes in patients with SJS compared with those in healthy controls (Figure 2). In addition, HMGB1 in the epidermis in SJS cases was detected diffusely though the epidermis compared with the findings in healthy donors, suggesting that HMGB1 is present in the extracellular spaces in SJS and/or TEN.

The role of HMGB1 detected in SJS and/or TEN is unknown. Since HMGB1 serves as an activator of the inflammatory cascade and to attract a wide variety of cells,^2,3 HMGB1 induced from necrolytic keratinocytes may act to stimulate inflammation and/or regeneration of the skin lesions. Increased levels of serum HMGB1 have been reported in several diseases such as severe infection and/or sepsis, trauma, cancers, and systemic lupus erythematosus,³ which were not observed in our patients with MPE, EM, and SJS and/or TEN.

It has been reported that granulysin and Fas ligand are possible candidates as biomarkers for early diagnosis of SJS and/or TEN,^5-7 but the duration of elevated granulysin and Fas ligand levels is limited^6,7; therefore, false-negative results for SJS and/or TEN could occur. In this regard, HMGB1 levels were high at the early stage of SJS and/or TEN and remained elevated even after the onset, which is in contrast to the kinetics of granulysin and Fas ligand. Although the numbers of patients with SJS and/or TEN were limited in this study, we propose that measurements of HGMB1 in combination with granulysin and/or Fas ligand would be a useful diagnostic tool for cases of SJS and/or TEN that require early diagnosis and treatment.

Correspondence: Dr Kabashima, Department of Dermatology, Kyoto University Graduate School of Medicine, 54 Shogoin-Kawara, Sakyo-ku, Kyoto 606-8507, Japan (kaba@kuhp.kyoto-u.ac.jp).

Accepted for Publication: April 13, 2011.

Author Contributions: Drs Nakajima and Kabashima had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Watanabe, Sugita, Hashimoto, Tokura, Miyachi, and Kabashima. Acquisition of data: Nakajima, Tohyama, Iijima, Nishimura, Doi, and Tanioka. Analysis and interpretation of data: Nakajima, Watanabe, Hashimoto, and Kabashima. Drafting of the manuscript: Nakajima, Watanabe, Tohyama, Sugita, Hashimoto, Tokura, Nishimura, Doi, Miyachi, and Kabashima. Critical revision of the manuscript for important intellectual content: Nakajima, Iijima, Tanioka, Miyachi, and Kabashima. Obtained funding: Tanioka and Kabashima. Administrative, technical, and material support: Nakajima, Watanabe, Tohyama, Sugita, Hashimoto, Nishimura, Doi, and Miyachi. Study supervision: Iijima, Hashimoto, Tokura, and Kabashima.

Funding/Support: This work was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology, and Health and Labor Sciences Research Grants (Research on Intractable Diseases) from the Ministry of Health, Labor, and Welfare of Japan.