Unusual gingival actinomycosis post allogeneic hematopoietic stem-cell transplant: case report

Twenty-one days after the last clinical follow-up, the patient presented a necrotizing lesion in the papilla region between the first and second molars of the second quadrant and no involvement of the rest of the hard palate or alveolar ridge. A biopsy was performed to confirm the diagnosis of an opportunistic infection in the region of the erythematous border and necrotizing area (Fig. 2).

Histopathological analysis showed ulcerated mucosa with dense lymphoid infiltrate with intermediate size lymphocytes, fibrosis, some plasma cells, and rare eosinophils (Fig. 3A). The presence of a bacterial cluster consistent with Actinomyces infection was also observed (Fig. 3B, C). Immunohistochemistry for CD20, CD3, and CD56 was performed to exclude the possibility that the lesion in the gingiva was a recurrence of extranodal NK/T cell lymphoma. There was a mixed infiltrate of mature B and T lymphocytes without atypia. As evaluated by in situ hybridization, the tissue was negative for Epstein-Barr virus (Fig. 4). After the histopathological analysis results, the patient was treated with amoxicillin (875 mg)-potassium clavulanate (125 mg), every 12 h for 4 weeks. There is no photographic record of the healed oral mucosa. The patient died due to aggressive lymphoma recurrence 93 days after the Actinomycosis diagnosis.

This patient was enrolled in a research protocol for the longitudinal profiling of the oral microbiota during allo-HSCT (Research Ethics Committee-Hospital Sírio-Libanês: #HSL 2016-08). Oral mucosa (OM), dental biofilm (DB), and gingival crevicular fluid (GCF) samples were collected at: D-5 (preconditioning), D5 (aplasia), D21 (engraftment), D50 (~ 4 weeks after engraftment), and D85 (~ 9 weeks after engraftment). Samples were processed and sequenced for microbiota profiling as described previously [8, 9]. Oral microbiota profiling showed drastic diversity changes at all oral sites during allo-HSCT with higher decrease in OM, but similar diversity levels were maintained across all oral sites during follow-up (Fig. 3D). As expected, Actinomyces was detected at high relative abundance in all oral sites before allo-HSCT, but abundance decreased during transplant. In the last sample analysed, there was an increase in Actinomyces abundance at all oral sites, particularly in GCF. Actinomyces relative abundance in the gingival crevicular fluid was 890% higher in D85 than in D-5, while other oral sites showed slight decreases compared to pre-allo-HSCT (− 20% in DB and − 49% in OM). The increase in Actinomyces abundance in GCF preceded the appearance of oral lesions by 20 days. Noteworthy, a single sequence variant of Actinomyces (100% identity with Actinomyces oris, Actinomyces naeslundii, and Actinomyces viscosus) was mainly responsible for this abundance increase, suggesting the involvement of a single or a few closely related opportunistic species.