Article Text

Download PDFPDF

Drug-induced lupus erythematosus following immunotherapy with anti-programmed death-(ligand) 1
  1. Jean-Marie Michot1,2,
  2. Mathilde Fusellier1,
  3. Stephane Champiat1,3,
  4. Charles Velter4,
  5. Capucine Baldini1,3,
  6. Anne-Laure Voisin5,
  7. Francois-Xavier Danlos1,
  8. Yolla El Dakdouki1,
  9. Maxime Annereau6,
  10. Xavier Mariette7,
  11. Caroline Robert4,
  12. Khadija Cherif8,
  13. Aurélien Marabelle1,
  14. Christine Mateus4,
  15. Olivier Lambotte2,3,9,10
  1. 1 Medical Oncology and Drug Development Department, Institut Gustave Roussy, Villejuif, France
  2. 2 Department of Internal Medicine and Clinical Immunology, Hôpital Bicêtre, Le Kremlin-Bicêtre, France
  3. 3 University of Paris Sud, Le Kremlin-Bicêtre, France
  4. 4 Department of Medical Oncology, Institut Gustave Roussy, Villejuif, France
  5. 5 Department of Pharmacovigilance, Institut Gustave Roussy, Villejuif, France
  6. 6 Department of Pharmacy, Institut Gustave Roussy, Villejuif, France
  7. 7 Department of Rheumatology, Hôpital Bicêtre, Le Kremlin-Bicêtre, France
  8. 8 Department of Biopathology, Institut Gustave Roussy, Villejuif, France
  9. 9 INSERM U1184, Immunology of Viral Infections and Autoimmune Diseases, Le Kremlin- Bicêtre, France
  10. 10 Commissariat à l’Energie Atomique (CEA), Fontenay-aux-Roses, France
  1. Correspondence to Dr. Jean-Marie Michot, Drug Development Department, Institut Gustave Roussy, Villejuif F-94805, France; jean-marie.michot{at}

Statistics from

We read with interest the study of Kostine et al describing rheumatic immune-related adverse events (irAE), which occur in 6.6% of patients treated for cancer by anti-programmed death-(ligand) 1 (PDL1).1 These new adverse effects pose significant challenges to patient care in terms of optimal management of these autoimmune damaging toxicities, while allowing effective antitumor therapy to continue.

The PD(L)1 pathway is involved in the maintenance of immune tolerance, and the blockage of this axis by anticancer immunotherapy could trigger autoimmune diseases and especially lupus.2 3 We then searched in the pharmacovigilance register of our institution—the ‘Registre des Effets Indésirables Sévères des Anticorps Monoclonaux Immunomodulateurs en Cancérologie (REISAMIC)’—whether cases of drug-induced lupus erythematosus (DI-LE) were reported following anti-PD(L)1 immunotherapies.

Between October 2013 and July 2017, five cases of DI-LE were recorded in REISAMIC. Given the number of patients having received anti-PD(L)1 during the same period (n=1044), the estimated incidence of DI-LE was 0.48%. All patients gave their written informed consent for the use of their data in this report. The patients’ characteristics are summarised in table 1. The patients had developed DI-LE at a median (range) age of 63 (48–80) years. None of the patients had a history of autoimmune disease before starting anti-PD(L)1. The most specific sign of DI-LE was subacute cutaneous lupus erythematous (SCLE) in four patients and chilblain lupus in the remaining patient. One patient having SCLE had also declared a systemic lupus erythematosus (SLE) according to the Systemic Lupus International Collaborating Clinics criteria.4 The DI-LE was revealed by a frank maculopapular rash in the four patients with SCLE (figure 1). The median time of DI-LE occurrence was 10 (range: 4–22) weeks after the initiation of immunotherapy. Antinuclear antibodies in serum were found positive for two (40%) out of the five patients and were specifically positive for anti-Sjögren’s syndrome-related antigen A (SSA). These two SSA-positive patients had SCLE but no eye or mouth dryness symptoms suggestive of Sjögren’s disease. A skin biopsy was performed in all cases except the chilblain lupus. The skin biopsies revealed a lymphocytic infiltrate of the dermis, predominantly around adnexal sites. Alcian blue staining revealed mucin deposits in all patients. Direct immunofluorescence assays for IgG or C3 in skin biopsy were positive in two of the four patients tested (50%). The treatment of DI-LE was based on topical corticosteroids in all cases, with the antimalarial hydroxychloroquine added in the SLE case, and the outcome was favourable with a resolution in all cases.

Table 1

Characteristics of the patients having developed DI-LE following treatment with anti-PD(L)1 immunotherapy

Figure 1

Photographs and histologic assessment of skin biopsies of cutaneous lupus erythematosus lesions induced by treatment with anti-PD(L)1. (A) Patient 4, erythematous papules and plaques with an annular, polycyclic configuration: generalised subacute lupus erythematous. (B) Patient 1, erythematous macules on the neck: subacute cutaneous lupus erythematosus. (C) Patient 2, symmetric papulosquamous erythematous rashes on the lower limbs. (D) Patient 3, erythematous macules and plaques on the back: subacute cutaneous lupus erythematosus. (E) Skin biopsy from patient 1, haematoxylin eosin saffron (HES) staining, ×2.5: peripheral and periadnexal monomorphic lymphocytic inflammatory infiltrate over the entire dermis. (F) Skin biopsy from patient 4, HES staining, ×5: lichenoid dermatosis with staged apoptotic bodies in the epidermis. Peripheral inflammatory mononuclear infiltrate in the upper dermis. (G) Skin biopsy from patient 3, Alcian blue staining, ×10: mucin deposits in the dermis.

This report is the first series of cases of lupus erythematosus induced by anti-PD(L)1 immunotherapy. A recent similar case report of pembrolizumab-related subacute cutaneous lupus erythematosus was provided.5 The DI-LE has been variously reported after drug exposure such as hydralazine, procainamide, quinidine, oestrogen, tumour necrosis factor inhibitors, chlorpromazine, isoniazid, practolol, penicillamine and minocycline.6 We believe that anti-PD(L)1 immunotherapies should also now be added to this list.

Based on our experience and the present case series, DI-LE induced by anti-PD(L)1 was characterised by an extensive, non-itchy and frankly macular or papular erythematous rash. The DI-LE diagnosis relies on the combination of the dermatological presentation associated with pathological features characterised by a lymphocytic dermal infiltration predominantly located at periadnexal sites, and mucin deposits.7 The confrontation between the clinical appearance and the pathological aspects is often useful to differentiate between DI-LE and other non-specific cutaneous irAEs, or other specific autoimmune skin diseases that can be induced by anti-PD(L)1 such as psoriasis, toxic epidermal necrolysis, lichen planus, bullous dermatitis and dermatomyositis.8

These new cases of lupus induced by anti-PD(L)1 should incite rheumatologist and internists to dedicate further prospective study for irAE. Investigation of potential biomarkers of irAEs such as the genetic background, serum levels of autoimmune factors and cytokines may help better understand these immunological adverse events and autoimmune conditions in general.


The authors thank David Fraser (Biotech Communication SARL, Ploudalmézeau, France) for copy-editing assistance.



  • Handling editor Josef S Smolen

  • Contributors All authors contributed to the patient care management and manuscript writing. All authors approved the manuscript submitted.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval Ethics Board Committee and Institutional Board of Institut Gustave Roussy.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles