Long-term cerebral atrophy and cognitive function in patients recovering from progressive multifocal leukoencephalopathy: two case reports

Abstract

Background

Long-term brain atrophy and cognitive function have rarely been reported in patients without pre-existing brain disorders who survived progressive multifocal leukoencephalopathy (PML).

Case presentation

Patient 1 was a 38-year-old woman with late-onset combined immunodeficiency, who presented with progressive cognitive decline and right hemiparesis. Brain MRI revealed a hyperintense lesion in the left frontal lobe. Patient 2 was a 53-year-old man in remission from follicular lymphoma, who experienced cognitive deterioration and displayed hyperintense signals in the left frontal and bilateral posterior lobes on MRI. Brain biopsy confirmed PML in both patients. After recovering from acute neurological decline, the cognitive and physical function of the patients was followed and we retrospectively analyzed sequential alterations in the MRI over eight years. Hyperintense lesion areas, cerebral hemisphere size, and intracranial volume were measured to calculate the monthly change ratios of the brain (MCRB), along with assessment of the third ventricle width. Both patients experienced no relapse. Wechsler Memory Scale-Revised scores improved in Patient 1 but slightly declined in Patient 2 despite stable Mini-Mental State Examination scores in both patients. MRI results were classified as lesion expansion, lesion reduction, and chronic atrophy. During the chronic atrophic stage, MCRB decreased by 0.06% and 0.09% in Patients 1 and 2, respectively. The third ventricle widened by 3 and 25 μm/month in Patients 1 and 2, respectively.

Conclusion

Progressive cerebral atrophy occurs during PML’s chronic stage, even without relapse. In these two cases, the patient with extensive bilateral lesions in the acute phase showed greater third ventricle enlargement and cognitive decline.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12883-026-04949-8.

Background

Progressive multifocal leukoencephalopathy (PML) is a demyelinating encephalopathy caused by the reactivation of the intracranial JC virus (JCV) in immunosuppressed patients, such as those with human immunodeficiency virus (HIV) or hematological malignancies. The mortality rate of PML is > 20% in patients who are HIV-negative, as reported in a recent Japanese study [1]. Given the rarity of the disease and its high mortality, little is known about long-term PML survivors. Although several reports have described patients with and without HIV infection who survived beyond the acute phase, data on long-term outcomes remain limited [2, 3].

Few previous studies have described long-term outcomes in patients with PML. One study, in which all but one patient was HIV-positive, reported that patients who survived for more than five years did not become bedridden, although they demonstrated subcortical atrophy of the brain [2]. Another study reported that 55% of patients with multiple sclerosis (MS) and natalizumab-associated PML exhibited worsened expanded disability status scales following treatment for PML [4]. However, most patients in these studies had a history of disorders impacting cognitive function, such as HIV infection and MS [5–7]. Therefore, the chronic influence of PML development in patients without underlying conditions affecting cognitions remains unclear.

In the present report, we presented two cases of PML who had no brain-affecting disorders and survived for more than eight years. Both developed progressive cerebral symptoms and brain biopsy resulted in a diagnosis of PML. We additionally conducted quantitative measurements of brain atrophy and cognitive function. We examined brain atrophy by measuring the brain area and calculated the width of the third ventricle (WTV), which was established to estimate whole-brain atrophy in MS studies [8].

Case presentation

Patient 1

A 38-year-old female with late-onset combined immunodeficiency and autoimmune hemolytic anemia treated with prednisolone presented with motor-predominant aphasia and right hemiparesis for 3 months. She was negative for HIV. An initial brain MRI revealed a hyperintense signal in the left frontal lobe (Fig. 1A). The initial JC viral load in the cerebrospinal fluid (CSF) was 4,000 copies/mL. Brain biopsy revealed oligodendrocytes with enlarged nuclei and intranuclear inclusions. Immunohistochemistry assays detected positive expression of the JC viral proteins, VP1 and VP2/3. Four months after symptom onset, oral prednisolone was reduced, and treatment with mirtazapine at a dose of 15 mg/day and mefloquine at a dose of 275 mg/day orally for three days, followed by 275 mg once per week was initiated. MRI performed 1.3 months after starting mefloquine revealed an enlarged fused lesion (Fig. 1B), which expanded into the left frontal and parietal lobes, while punctate lesions were observed in the right hemisphere. No fused lesions or obvious thalamic involvement were detected bilaterally. Subsequent MRI showed a gradual shrinkage of the lesions (Fig. 1C), and 5 months after onset, the JC viral load in the CSF was undetectable. The patient’s aphasia improved, and she was able to walk unassisted by 9 months post-onset. Mefloquine was discontinued 26 months after disease onset, and no neurological deterioration was observed. Cognitive function showed improvements in Wechsler Memory Scale-Revised (WMS-R) subcategories (logical memory, visual memory, general memory, and delayed recall), as well as in the Single Digit Modalities Test (SDMT), and Paced Auditory Serial Additions Task Clinical Assessment for Attention (PASAT) tests (Table 1). MRI taken 74 months after PML onset demonstrated reduced lesions and diffuse atrophy of the left cerebral hemisphere (Fig. 1D). Transient reappearance of JCV in the CSF was observed 97 months after PML onset, without any associated clinical deterioration. Additional MRI examinations were performed between these time points based on the clinical indications. All axial fluid-attenuated inversion recovery (FLAIR) slices showing visible lesions at disease onset, at the acute phase (when the lesion reached its maximum size), and at final follow-up are provided in Additional File 1.

Fig. 1.

Sequential brain MRI figures of the patients. Sequential alterations in the brain magnetic resonance images of Patient 1 (A-D) and Patient 2 (E-H). These images were assessed before treatment (A and B for Patient 1; E and F for Patient 2), at the lesion reduction stage (C and G), and at the chronic atrophic stage (D and H)

Table 1.

Long-term follow up of cognitive function in Patient 1 and Patient 2

Months from the PML onset		Patient 1		Patient 2
		29	62	17	127
MMSE		29	30	25	24
WMS-R	Logical memory	99	109	85	71
	Visual memory	87	108	70	72
	General memory	95	109	77	67
	Attention	115	111	82	62
	Delayed recall	103	109	72	64
SDMT		56.4	65.4	10	15.5
PASAT (2 s)		85	98	2	Not completed

PML progressive multifocal leukoencephalopathy, MMSE Mini-Mental State Examination, WMS-R Wechsler Memory Scale-Revised, SDMT Single Digit Modalities Test, PASAT Paced Auditory Serial Additions Task Clinical Assessment for Attention

Patient 2

A 53-year-old male with complete remission of follicular lymphoma experienced memory disturbance, apraxia, and right-sided unispatial neglect for 2 months. Test results for HIV were negative. Initial brain MRI revealed hyperintense signals in the bilateral posterior and left frontal lobes (Fig. 1E). The initial JC viral load in the CSF was not detected. Brain biopsy from the left parietal lobe revealed oligodendrocytes with dot-shaped intranuclear inclusions and immunohistochemistry assay with antibodies to VP2/3 was positive. Treatment with risperidone at a dose of 1 mg/day, mirtazapine at a dose of 15 mg/day, and mefloquine at a dose of 275 mg/day orally for three days, followed by 275 mg once per week was initiated. The lesions expanded until 1.7 months post-treatment (Fig. 1F), though no lesions were noted in the thalami. Cognitive dysfunction subsequently improved, and hyperintense lesions on FLAIR images diminished (Fig. 1G). The patient demonstrated no acute neurological decline, and he returned to work as a dance teacher. Neuropsychological tests revealed that the Mini-Mental State Examination (MMSE) score remained stable (25 points at 17 months, and 24 points at 127 months, after the onset of PML), and the SDMT scores were higher at 127 months than at 17 months. However, responses to the WMS-R suggested decreased attention and short-term memory (Table 1). Neither lesion expansion nor new lesions were observed on MRI over a ten-year period, though brain atrophy progressed slowly (Fig. 1H). Repeated CSF studies revealed no reappearance of JCV. Additional MRI examinations were performed between these time points based on the clinical indications. All axial FLAIR slices showing visible lesions at disease onset, at the acute phase (when the lesion reached its maximum size), and at final follow-up are provided in Additional File 1.

MRI assessment

We measured the area of the hyperintense lesion (AL) in the section where the lesion was the largest, and the areas of the cerebral hemisphere (AB) and intracranium (AI) in the same section. The monthly changing ratios for the lesion (MCRL) and brain (MCRB) were defined as the changes in the relative areas of AL to AI and AB to AI, respectively. Measurements were obtained from axial FLAIR images using an image analysis software (Mango [9]). The WTV was measured in the axial T1-weighted images according to previously published methods employed in a study of patients with MS [7]. Briefly, we obtained the WTV by drawing a line through the long axis of the ventricle parallel to the interhemispheric fissure in the section where the third ventricle was most visible. Width was measured by drawing a second line perpendicular to the first line at its midpoint. The change ratios of WTV were subsequently calculated. This procedure is summarized for clarification in Fig. 2. Additionally, the area ratios (AL/AI and AB/AI) were calculated across lesion-containing slices consistently available across time points and averaged to obtain mean lesion and brain ratios of the whole slices (mean AL/AI and mean AB/AI), confirming overall tendency consistent with primary single-slice measurements.

Fig. 2.

Schematic visualization of MRI measurements. A The relative lesion area was defined as the ratio of the hyperintense lesion area (AL) to the intracranial area (AI). The relative brain area was defined as the ratio of the cerebral hemisphere area (AB) to AI. The monthly changing ratios for the lesion (MCRL) and brain (MCRB) were then calculated based on sequential changes in these relative areas. B Measurement of the width of the third ventricle (WTV). On axial T1-weighted images, a line was drawn along the long axis of the ventricle parallel to the interhemispheric fissure in the slice where the third ventricle was the most visible. The width was determined by a perpendicular line crossing the midpoint of the long axis

We classified the patients’ course into three stages based on the MCRL on MRI: lesion-expansion stage (monthly change > 0%), lesion-reduction stage (monthly change < -1%), and chronic atrophic stage (monthly change between − 1% and 0%) (Fig. 3A, B). In the lesion-expanding stage, the average MCRL and MCRB values of Patient 1 and Patient 2 were 6.32% and 1.65% per month, respectively. In the lesion reduction stage, after the initiation of treatment, the average MCRL and MCRB values of both patients were − 3.29% and − 1.16% per month, respectively. In the chronic atrophic stage, the average MCRL and MCRB values of Patient 1 and Patient 2 were − 0.11% and − 0.08% per month, respectively (Table 2). Figure 3 shows the sequential changes in MCRL and MCRB.

Fig. 3.

Long-term quantitative alterations in brain morphology in PML. Long-term changes in the cross-sectional areas of the brain and PML lesion and width of the third ventricle in Patients 1 (A) and 2 (B). In the upper graph, the vertical axis demonstrates the ratio of the brain area to the intracranial area (solid line) and the ratio of the PML lesion area to the intracranial area (dotted line). In the lower graph, the vertical axis illustrates the width of the third ventricle. PML: progressive multifocal leukoencephalopathy

Table 2.

Monthly change ratio of the relative area of the PML lesion to the intracranial area (MCRL), relative area of the brain parenchyma to the intracranial area (MCRB), and width of the third ventricle

Monthly change ratio		Lesion-expansion stage	Lesion-reduction stage	Chronic atrophic stage
Relative area of the lesion to the intracranial area (MCRL; %/month)	Patient 1	6.8	-2.0	-0.1
	Patient 2	5.8	-4.6	-0.1
	Average	6.3	-3.3	-0.1
Relative area of the brain parenchyma to the intracranial area (MCRB; %/month)	Patient 1	1.6	-0.6	-0.1
	Patient 2	1.7	-1.7	-0.1
	Average	1.7	-1.2	-0.1
Enlargement of the third ventricle (µm/month)	Patient 1	-369	675	3
	Patient 2	-111	376	25
	Average	-240	526	14

MCRL and MCRB were subdivided into three phases: lesion-expanding stage (MCRL more than 0%), lesion reduction stage (MCRL less than − 1%), and chronic atrophic stage (MCRL between 0 and − 1%). PML: progressive multifocal leukoencephalopathy

The monthly change ratios of WTV in Patient 1 and Patient 2 were − 369 and − 111 μm per month in the lesion-expansion stage, 675 and 376 μm per month in the lesion reduction stage, 3 and 25 μm per month in the chronic atrophic stage, respectively (Table 2). A multislice analysis performed at representative time points showed a similar tendency in each patient, with mean lesion ratio (AL/AI) peaking at maximal lesion extension and decreasing at the final follow-up, whereas mean brain ratio (AB/AI) at the final follow-up were lower than those at baseline (Additional file 1).

Discussion

Herein, we presented a sequential analysis of brain atrophy and cognitive function in two patients with PML who survived for more than eight years. A progressive decline in brain volume was noted in both patients; however, cognitive functions were not necessarily disturbed.

Brain atrophy, indicated by an area reduction in the cerebral hemisphere and enlargement of the third ventricle, progressed for several years after remission of PML. MRI analysis showed an expansion and subsequent reduction of the brain parenchymal area, similar to previous case studies that used quantitative or semi-quantitative measurements [10, 11]. Subcortical atrophy was reported after several years of intensive PML treatment in stable patients in the chronic phase of the disease [2]. Patients in this study demonstrated continuous cerebral hemisphere atrophy, which provides insights into the pathophysiology of the PML remission phase. In long-surviving patients with post-acute PML, the common findings in previous pathological studies highlight chronic myelin damage, probably caused by JCV-infected oligodendrocyte destruction and not by swollen nuclei [12–15]. This implies that the chronic intracellular existence of JCV can be prolonged; however, viral reproduction, as suggested by swollen nuclei [16], may be prevented by brain immune system recovery. Notably, JCV infection in mice with humanized glial precursor cells limited glial precursor cell differentiation and inhibited remyelination [17]. Thus, we speculate that dysfunction of the remyelinating ability, rather than active demyelination due to viral expansion, leads to progressive brain atrophy in PML survivors.

In the chronic atrophic stage, enlargement of the third ventricle, reflecting atrophy of the thalamus, was noted in Patient 2, but not in Patient (1) This difference might be attributed to the extent of the maximum fused lesion in the acute phase, which covered the frontal lobe in both patients but extended to the bilateral posterior lobe only in Patient (2) The enlargement of the third ventricle in Patient 2 was likely due to the network dysfunction resulting from the more extensive lesion rather than direct thalamic damage given the role of the thalamus as a connector hub of the cortical network [18] and the absence of acute thalamic lesions in both cases. Previous studies in patients with MS have reported that WTV is linked to total brain atrophy and progression at all MS stages [19]. The speed in the chronic atrophic stage was equivalent to 1.6 times faster in Patient 2 than the previously reported data of secondary progressive MS, in which the annual enlarging ratio of the WTV was 0.19 mm per year [20]; conversely, our data revealed a rate of 0.025 mm per month, corresponding to 0.3 mm per year. These findings suggest that the rate of increase in the WTV in the remission phase of PML correlates with the size and anatomical distribution of the maximal fused lesion in the acute phase.

In our study, neurological examination revealed preserved motor function in both patients, and the MMSE results indicated generally stable cognitive function. Patient 1 emphasized that once the acute phase of PML was overcome, cognitive function slowly improved despite progressive atrophy of the cerebral hemisphere. However, a more detailed assessment of the memory function using the WSM-R suggested that memory impairment slowly progressed in Patient 2, in whom the relatively larger acute PML lesion may have contributed to cognitive function. While the MMSE scores remained stable in both patients, suggesting minimal global cognitive decline due to aging, differences in specific neuropsychological domains likely reflect the combined effects of disease progression, lesion size, and follow-up duration, rather than overall cognitive deterioration. In addition, the influence of aging should also be considered as a potential confounder on cerebral atrophy. Although we were unable to recruit age-matched healthy controls for comparison, Patient 2 exhibited a faster rate of ventricular enlargement than that reported in MS, which also progresses faster than normal aging [21]. This finding suggests that the progression of brain atrophy in Patient 2 cannot be solely explained by normal aging. The different course of brain atrophy in our cases could also have been influenced by the age at PML onset. Further studies are necessary to elucidate the mechanisms underlying chronic progressive brain atrophy and cognitive functional variability in patients with post-acute PML.

This study has certain limitations. First, the analysis was based on only two cases and lacked age-matched healthy controls or patients with MS as a comparison group, limiting the generalizability of our findings. Second, MRI data were not obtained using volumetric methods, preventing precise measurement of total brain volume and lesion-specific atrophy. Consequently, the contribution of overall brain volume loss to cognitive function changes could not be fully assessed. Therefore, differences in neuropsychological outcomes between the two patients may reflect both the extent and location of PML lesions and the underlying variations in global brain volume, which were not quantified in this study. Longitudinal structural changes were evaluated using normalized area indices derived from a single representative MRI slice. A multi-slice analysis performed at representative time points showed a similar tendency, supporting the consistency of the primary single-slice measurements and suggesting that these indices may serve as surrogates for relative lesion and brain volume within the intracranial space. Future studies using larger cohorts and volumetric MRI techniques are required to clarify the relationship between regional and global brain atrophy and cognitive trajectories in long-term PML survivors.

Conclusion

Patients who recovered from PML in the chronic phase showed slowly progressive brain atrophy, whereas physical function was preserved. The long-term detailed cognitive outcome may be associated with fused lesions involving bilateral regions in the acute phase.

Abbreviations

PML

Progressive multifocal leukoencephalopathy

JCV

JC virus

HIV

Human immunodeficiency virus

MS

Multiple sclerosis

WTV

Width of the third ventricle

CSF

Cerebrospinal fluid

MRI

Magnetic resonance imaging

VP1

Viral protein 1

VP2/3

Viral protein 2/3

WMS-R

Wechsler Memory Scale-Revised

SDMT

Single Digit Modalities Test

PASAT

Paced Auditory Serial Additions Task

AL

Area of the lesion

AB

Area of the cerebral hemisphere

AI

Area of the intracranium

MCRL

Monthly changing ratio for the lesion

MCRB

Monthly changing ratio for the brain

FLAIR

Fluid-attenuated inversion recovery

MMSE

Mini-Mental State Examination

Authors’ contributions

All authors were involved in the management of this patient. Conception: NS. Data analysis: ST, DO. Writing the draft: ST, DO, NS. Supervision: TY, NS. All authors read and approved the manuscript.

Funding

This work was partly supported by a Grant-in-Aid from the Research Committee of Prion Disease and Slow Virus Infection of the Ministry of Health, Labour, and Welfare of Japan (NS).

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the principles of the Declaration of Helsinki (2013). The study protocol was approved by the Institutional Ethics Committee of Institute of Science Tokyo (M2000-1774).

Consent for publication

Written informed consent for publication of the present case report and accompanying images was obtained from each patient. A copy of the written consent is available for review by the Editor of this journal.

Competing interests

The authors declare no competing interests.