Introduction

Paraneoplastic pemphigus (PNP), an autoimmune disorder that can arise from malignancies, is fatal when complicated with bronchiolitis obliterans (BO), similar to that in patients receiving hematopoietic stem cell transplantation (HSCT). Autoantibodies to the desmosome proteins envoplakin and periplakin of epithelial cell adhesion molecules are essential for diagnosing PNP. BO after HSCT is an unfavorable complication that develops in association with chronic graft-versus-host disease (cGVHD). The effectiveness of rituximab on BO has been reported, which suggests that BO after HSCT is associated with autoimmunity similar to PNP.

Here we analyzed anti-envoplakin and anti-periplakin antibodies that are associated with PNP in patients with and without BO complications who developed cGVHD after HSCT.

Patients and Methods

We analyzed anti-envoplakin and anti-periplakin antibodies in 21 patients (11 males and 10 females) with (n = 10) or without (n = 11) BO complications who developed cGVHD following HSCT between 1988 and 2015 at our institution. The median age at the time of HSCT in patients with and without BO was 10.8 (range, 1-21) and 7.5 (range, 1-14) years, respectively. The median duration from HSCT to BO onset was 261.6 (range, 41-825) days. We analyzed anti-envoplakin and anti-periplakin antibodies in cryopreserved sera collected before and after BO or cGVHD onset using immunoprecipitation (IPP) and enzyme-linked immunosorbent assay (ELISA) for detecting both autoantibodies. Longitudinal sera were obtained from a patient with cGVHD. Fifty-eight serum samples were collected from 21 patients. Biotinylated recombinant periplakin and envoplakin were produced from cDNAs using the transcription and translation (TnT) T7 Quick Coupled Transcription/Translation System (Promega) with rabbit reticulocyte lysates. IPP and ELISA were performed using in vitro TnT products as previously reported (Muro Y, et al. Rheumatology 2012;51:1508-13).

Results

On SDS-polyacrylamide gel electrophoresis, biotinylated recombinant periplakin and envoplakin showed protein bands with predicted size of 190 kDa and 210 kDa, respectively. The serum samples from the PNP patient and normal control serum were examined in an ELISA titration study. These titration curves showed the possibility of quantitatively measuring these autoantibodies. Both anti-envoplakin and anti-periplakin antibodies were positive in two of 10 patients with BO at BO onset but negative in all 11 patients without BO during cGVHD. In one of the two patients with plakin family autoantibodies, titers of anti-periplakin and anti-envoplakin antibodies gradually decreased after administering rituximab five times and repeating plasmapheresis seven times and subsequently decreased to below cut-off level. Although oxygenation substantially decreased consistent with the clearance of autoantibodies, the patient died of invasive pulmonary aspergillosis during steroid tapering.

In the other patient, plakin family autoantibodies in the sera became negative after administering rituximab nine times. The patient is alive, and the spirometry findings were normal 6 years after the clearance of both autoantibodies.

Conclusions

Our findings indicate that plakin family autoantibodies may participate in BO onset in some patients receiving HSCT. The detection of these autoantibodies in patients with BO may be a good biomarker of cGVHD, providing a rationale and target for repeated rituximab therapy.

Disclosures

Kojima:SANOFI: Honoraria, Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution