Polyomaviruses are circular double-stranded DNA (dsDNA) viruses. Thirteen polyomaviruses are classified as human polyomaviruses (HPyV); eleven routinely infect humans, however only six are associated with human diseases, the most relevant being BK polyomavirus (BKPyV), JC polyomavirus (JCPyV) and Merkel cell polyomavirus (MCPyV) (Cook, 2016).

BKPyV (HPyV1) is known as the etiological agent of polyomavirus nephropathy in kidney transplant patients and of hemorrhagic cystitis in hematopoietic stem cell transplant (HSCT) recipients (Aldiwani et al., 2019; Kant et al., 2022; McIlroy et al., 2019). MCPyV is responsible for a form of Merkel cell carcinoma (Feng et al., 2008).

JC virus is commonly associated with central nervous system (CNS) tropism, causing a progressive multifocal leukoencephalopathy (PML) (Haley and Atwood, 2017), a serious adverse event for some immunosuppressive therapies (Cook, 2016; Cortese et al., 2021).

Immunosuppression plays a crucial role in BKPyV and JCPyV pathology, leading to the reactivation of the infection, although with great heterogeneity ranging from asymptomatic replication to organ dysfunction such as polyomavirus nephropathy, or even a lethal outcome such as PML. Only in a minority of cases BKPyV and JCPyV reactivations lead to overt pathology (Hirsch et al., 2002).

In immunosuppressed patients, BKPyV reactivation commonly leads to nephropathy or hemorrhagic cystitis, less frequently BKPyV has been reported to cause fatal pneumonia, retinitis, native kidney nephritis (Brennan et al., 2005).

BKPyV meningoencephalitis in severely immunocompromised patients is a very rare event, and BKPyV tropism for the brain is poorly understood (Friedman and Flanders, 2006). Few cases of BKPyV encephalitis were reported in HSCT recipients (Friedman and Flanders, 2006; Yamaguchi et al., 2024).

A study detected in the cerebrospinal fluid (CSF) of children with clinically diagnosed meningitis and/or encephalitis of unknown etiology, BKPyV and MCPyV in two and four samples, respectively (Launes et al., 2024).

In a study of 2,062 CSF samples from patients with neurological disorders, BKPyV was detected in 20 patients diagnosed with PML or multiple sclerosis (Bárcena-Panero et al., 2012). CSF samples from BKPyV-positive patients were also screened for other neurotropic viruses, including JCPyV, herpes simplex virus (HSV), Epstein–Barr virus (EBV), cytomegalovirus (CMV), and human enteroviruses (EVs). Among these, one patient tested positive for JCPyV, one for HSV, one for EBV, and one for both EBV and CMV.

Mycosis fungoides is the most common form of cutaneous T-cell lymphoma and can progress to advanced stages such as the erythrodermic variant, which is often resistant to conventional therapies. In recent years, Mogamulizumab, a monoclonal antibody targeting the receptor C-C chemokine receptor 4 (CCR4) receptor, has emerged as a novel therapeutic option, although associated with a risk of immunosuppression (Whittaker et al., 2016).

Herein we present a fatal case of PML associated with BKPyV in a 66-year-old male patient diagnosed with erythrodermic mycosis fungoides receiving Mogamulizumab.

Few cases of BKPyV encephalitis have been reported in HSCT patients and in other immunodeficient conditions (Jun et al., 2016; Yamaguchi et al., 2024). In BKPyV encephalitis a variety of symptoms have been described including weakness, vision problems, speech difficulties, and cognitive changes, in our case, progressive cognitive and motor deficits were observed. Usually, computed tomography (CT) and MRI are nonspecific, as in our case.

In a study the diagnosis of BKPyV encephalitis was based on the detection of BKPyV DNA PCR in CSF, complemented by BKPyV PCR biopsy specimen (Stoner et al., 2002). In our case a brain biopsy was not performed due to the clinical conditions of the patient.

Based on the clinical picture, of the positive result for BKPyV PCR in CSF, and MRI suggesting primarily an inflammatory pathology it is reasonable to consider this case as a BKPyV-associated PML.

It is important to note that BKPyV DNA was detected in three different CSF samples obtained from lumbar puncture and in plasma lower value of BKPyV DNA were observed, allowing exclusion of blood contamination of the CSF. In our case BKPyV DNA was detected at levels similar to those observed in previously reported cases (10^2–10{sp}3{it} {/sp}{/it}copies/mL), although the specificity and optimal cut-off values of BKPyV DNA in the CSF are unknown and clear diagnostic criteria for BKPyV encephalitis need to be established (Stoner et al., 2002; Chittick et al., 2013). Serological testing showed high anti-BKPyV and anti-JCPyV IgG levels in plasma, indicating prior exposure to both viruses, while antibodies were undetectable in CSF, consistent with lack of intrathecal synthesis. This compartmental dissociation in antibody distribution between plasma and CSF, along with the detection of BKPyV DNA and absence of JCPyV DNA in CSF, supports the hypothesis of BKPyV neurotropism. The lack of antibodies in CSF, combined with the detection of BKPyV DNA and the absence of JCPyV DNA, suggests that BKPyV was the causative agent of the PML. Although these findings support the interpretation of viral reactivation, definitive proof would require documentation of BKPyV infection prior to disease onset; therefore, primary infection occurring during profound immunosuppression cannot be completely excluded. In addition, because neither brain biopsy nor post-mortem examination were available, direct histopathological confirmation and viral detection within CNS lesions could not be performed, precluding definitive etiological attribution at the tissue level.

Our sequencing results highlight a clear intra-host variability of BKPyV across compartments, involving both VP1 (Helle et al., 2017) and the NCCR (Walder et al., 2025). At the VP1 level, we observed conserved genotype II assignment, in line with previous reports showing genotyping stability despite minor sequence heterogeneity (McIlroy et al., 2019; Morel et al., 2017). Four major genotypes of BKPyV are known: I, II, III and IV. Genotype I is the most prevalent and distributed worldwide, genotype IV is present in East Asia and Europe, whereas genotypes II and III are rarely observed; notably, genotype II has also been reported in Europe at a low frequency (Blackard et al., 2020; Jin et al., 1993; Yogo et al., 2009). In this context, the patient was born and raised in Italy, with no known ethnic or geographic background differing from the general Italian population. To date, no clear correlations between BKPyV genotype and pathology have been established and further studies are required to assess a neurotropic role for genotype II and in PML development.

The major capsid protein of BKPyV, VP1, is involved in the interaction with host cellular receptors but whether specific VP1 polymorphisms are linked to neurological sequelae of BKPyV is not clear (Neu et al., 2009; Saribas et al., 2019). The detection of compartment-specific changes in the CSF, such as the G57 insertion and N62K, is consistent with intra-patient VP1 evolution and may represent early signatures of selective pressure (McIlroy et al., 2019). This observation aligns with findings by Darbinyan et al., 2016, who reported genetic divergence of BKPyV in brain tissue from immunocompromised individuals, highlighting the potential for neurotropic variants to emerge in-vivo.

For the NCCR, we identified marked compartment-specific differences. NCCR rearrangements have been linked to altered transcriptional activity in vitro (Bárcena-Panero et al., 2012), although archetypal NCCRs have also been found in CSF isolates, indicating they are not universal markers of neurotropism. In kidney transplant recipients, rearrangements involving the Q and R blocks have been associated with enhanced viral replication and, in some cases, with more severe clinical manifestations (Virtanen et al., 2018; Liimatainen et al., 2020). In our patient, the CSF-derived isolate displayed a ΔQ–R NCCR variant, raising the possibility that similar structural alterations may facilitate viral adaptation or replication within the CNS. However, direct evidence supporting this mechanism in brain infection is currently lacking, and the functional relevance of these rearrangements in neurotropism remains to be determined. These findings suggest that BKPyV undergoes compartment-specific evolution, with VP1 mutations and NCCR rearrangements acting in parallel. This intra-host heterogeneity likely reflects the interplay of host immune pressure, replication dynamics, and tissue-specific selective forces. Further studies are needed to elucidate whether such compartmentalization influences viral pathogenicity, immune evasion, or response to therapy.

A genetic predisposition to JCPyV-induced PML is supported by the molecular diagnosis of inborn errors of immunity (IEIs) in PML patients. To date, 26 genes believed to cause IEIs and PML have been identified (Eis et al., 2025). Genetic studies are required to assess whether our patient was a carrier of a known genetic IEIs or whether an alternative genetic background might be associated with the BKPyV-induced PML. Overall, the baseline immunological vulnerability of the patients, with pre-existing bronchiectasis and history of recurrent infections, support the hypothesis of IEIs.

Mogamulizumab is a defucosylated humanized monoclonal antibody targeting the CC chemokine receptor 4 (CCR4), highly expressed on malignant T cells in cutaneous T-cell lymphoma and adult T-cell leukemia/lymphoma (Whittaker et al., 2016; Zengarini et al., 2024). Its antitumor activity is mediated primarily through enhanced antibody-dependent cellular cytotoxicity via FcγRIIIa engagement on natural killer cells, leading to efficient elimination of CCR4-positive malignant cells. In parallel, Mogamulizumab depletes CCR4-expressing regulatory T cells (Tregs), thereby modulating the host immune environment. In addition, blockade of the CCR4–CCL17/CCL22 signaling axis may impair lymphocyte trafficking and tissue homing. While these mechanisms support therapeutic efficacy, the profound reduction of Tregs and resulting immune imbalance may impair immune surveillance against latent viral pathogens. Accordingly, cases of viral reactivation and opportunistic infections, including hepatitis B virus reactivation and cytomegalovirus disease, have been reported during Mogamulizumab therapy (Wang et al., 2020). Our case suggests that a similar immune dysregulation may predispose to BKPyV reactivation and CNS dissemination, underscoring the importance of infectious monitoring in treated patients. In our case, for the first time in a Mogamulizumab-treated patient, JCPyV and BKPyV reactivations were observed, and BKPyV-associated PML was diagnosed, highlighting the increased risk of opportunistic infections. The profound lymphopenia induced by Mogamulizumab, together with Treg depletion, might predispose patients to unusual opportunistic infections, including neurotropic polyomaviruses.

Current Mogamulizumab protocols lack standardized guidelines for viral monitoring, however Mogamulizumab therapy carries a non-negligible risk of severe infections due to its immune-modulating mechanism. Our case showing the reactivation of CMV, JCPyV and BKPyV infections highlights the relevance of a closer surveillance for viral reactivations of CMV, JCPyV, BKPyV, and others such as EBV, particularly in high-risk patients. Moreover, our case suggests that BKPyV-associated PML may be an emerging, under-recognized complication in patients receiving potent T-cell-targeted therapies such as Mogamulizumab. The case underscores the need for heightened vigilance for atypical CNS viral infections, including BKPyV -associated PML, in patients treated with Mogamulizumab, especially in the presence of pre-existing immunodeficiency, since the immunological vulnerability of the patient may have further increased the susceptibility to a severe viral complication during therapy.

The monitoring could be helpful to understand whether Mogamulizumab plays a role in the development of neurotropic variants of BKPyV.

Further studies are warranted to better define prophylactic and monitoring strategies, including a genetic assessment for IEIs.

Finally, our case supports the importance of considering BKPyV infection in the differential diagnosis of immunosuppressed patients presenting with clinical symptoms of encephalopathy and undetectable JCPyV in the CSF.

In conclusion, BKPyV typically causes nephropathy in transplanted patients, although its neurotropic potential has been recognized in immunocompromised hosts. So far, rare cases of BKPyV-associated PML have been reported and our case is the first associated with Mogamulizumab treatment. This case highlights the potential role of BKPyV as a neurotropic pathogen in the setting of Mogamulizumab-induced immunosuppression. Although systematic BKPyV screening is not currently recommended, targeted testing should be considered in patients developing unexplained neurological deterioration under CCR4 blockade.