The boundaries between PML and PML-IRIS: difficult to define, pathology may predict Article type: Original Article

Jiamin Chen¹Yanni Du¹Lei Sun¹Yuyang Dai²Liang Zhang¹Xinghua Ding¹Ting Liu¹Kun Yang¹Xiaoyi Han¹Man Li¹Xingang Zhou^1*

• ¹Beijing Ditan Hospital, Capital Medical University, Beijing, China
• ²Beijing Tongren Hospital, Capital Medical University, Beijing, Beijing Municipality, China

: difficult to define, pathology may predict Background: Progressive multifocal leukoencephalopathy (PML), caused by John Cunningham (JC) virus reactivation, represents a critical neurological complication in AIDS-related immunosuppression. This single-center study conducted a clinicopathological analysis of 19 confirmed PML cases in an AIDS cohort (16 biopsy; 3 surgical specimens), employing comprehensive neuropathological evaluation. Immunohistochemical testing included SV40, NF, NeuN, P53, Ki-67, GFAP, Oligo-2, and CD68. Myelin architecture was evaluated through Luxol fast blue staining, complemented by molecular diagnostics incorporating quantitative JC viral load PCR and metagenomic next-generation sequencing (mNGS). Results: Notably, 63.2% (12/19) of them had blood CD4+ T-cell counts < 200 cells/μl, and 36.8% (7/19) had ≥ 200 cells/μl. 52.9% (9/17) of the patients had elevated CSF protein, 5.3% (1/19) had decreased CSF glucose. Statistical analysis revealed significant correlations between mass effect and both blood CD4+ T-cell counts (P = 0.022) and CSF protein levels (P < 0.001). It also demonstrated significant positive correlations between the duration of HIV diagnosis and the degree of inflammatory infiltration (P = 0.038) and perivascular inflammatory infiltration (P = 0.005), as well as plasma cell infiltration (P = 0.011). The degree of inflammatory infiltration was significantly positively correlated with antiretroviral therapy (ART) (P = 0.036). The degree of inflammatory infiltration, the presence of plasma cells, and perivascular lymphocytic cuffing were significantly associated with contrast enhancement on imaging studies (P = 0.044, P = 0.011, and P = 0.018, respectively). These cases display characteristics that deviate from the classic PML previously reported, exhibiting a tendency towards MRI enhancement and histologically indicating a more severe inflammatory response, especially for patients following ART treatment. Conclusion: Our findings suggest that PML and PML-immune reconstitution inflammatory syndrome (IRIS) represent a continuous pathological spectrum, potentially bridged by an intermediate stage with distinct clinicopathological features. This transitional phase may constitute a critical link in the continuum between classic PML and fully developed PML-IRIS. Importantly, it implicates synergistic mechanisms of viral oncogenesis and immune reconstitution, which could redefine therapeutic strategies for this emerging PML variant.