Human T-Lymphotropic Virus Type 1 and Cryptococcosis Infection, an Underdiagnosed Association: Case Series and Literature Review

Abstract Clinical and epidemiological features of 7 human immunodeficiency virus–negative Peruvian patients coinfected with human T-lymphotropic virus type 1 (HTLV-1) and cryptococcosis (2006–2017) were studied. Most cases had meningeal involvement, were male, and originated from Peru's jungle. Patients with cryptococcosis should be tested for HTLV-1 in endemic areas of this retrovirus.

Cryptococcosis causes 250 000 deaths per year and primarily affects people with HIV (PWH).In people without HIV, the disease predominantly affects the central nervous system and the lungs.Mortality attributable to cryptococcosis is higher in HIV-negative people than in PWH [10].
Peru is an endemic country for HTLV-1, reporting the higher incidence of infection in ethnic groups from Andean (Quechua, Aymara) [27,28] and from jungle (Shipibo-Conibo, Quechua-Lamas, Wampis, and Amarakaeri) regions [29].In a casecontrol study, 85.7% of Peruvians subjects with Strongyloides stercoralis hyperinfection had HTLV-1 infection [30].In addition, León et al reported 4 patients with HTLV-1 and severe Th2 immunosuppression presenting with disseminated paracoccidioidomycosis who acquired the fungal infection in the Peruvian jungle [31,32].However, although HTLV-1 is endemic in Peru, until now no description of HTLV-1 infection in subjects with cryptococcosis has been reported.Therefore, the aim of this study was to describe the clinical features of patients with cryptococcosis who tested positive for HTLV-1 in an endemic country for the latter.A literature review of this association was also performed.

METHODS
This study is a case series conducted at the Alexander von Humboldt Tropical Medicine Institute and Cayetano Heredia Hospital, in Lima, Peru, between January 2003 and December 2017, which serves as a referral hospital for HTLV-1 infection.We enrolled all non-HIV-infected patients with a diagnosis of cryptococcosis, defined as isolation of Cryptococcus spp in culture from a sterile site.Medical records of subjects with cryptococcosis were reviewed, and data from patients with an HTLV-1 result were analyzed.The fungal identification was performed by conventional methods (India ink stain, urea assimilation, growth on selective agar Guizotia abyssinica at 37°C) at the Tropical Institute Alexander von Humboldt.Cultures were considered negative if Cryptococcus spp were not detected at the fifth week of incubation.The L-canavanine-glycine-bromothymol blue Open Forum Infectious Diseases B R I E F R E P O R T medium was used to differentiate Cryptococcus gattii from Cryptococcus neoformans.HTLV-1 diagnosis was performed using a chemiluminescence test (6L61 Architect HTLV-I/II Reagent Kit, Abbott Diagnostics).The positive samples were confirmed by Western blot analysis.
We attempted to identify all studies regardless of language or publication status (published, unpublished, in press, and in progress) that had reported on the coinfection by these 2 pathogens.The following terms were used for the search: Cryptococcus, HTLV-1, and cryptococcal infections in non-HIV-infected populations.The relevant articles from January 1980 to March 2023 were searched at PubMed, Medline, and Embase databases.
The descriptive statistical analysis of clinical, demographic, and laboratory data were made using Stata 17 software (StataCorp, College Station, Texas).
The study protocol was evaluated and approved by the Institutional Research Ethics Committee of the Continental University (approval document number 0149-2023-CIEI-UC).

RESULTS
During the study period, 339 positive Cryptococcus spp culture samples were found in HIV-negative participants and PWH; 38 of them occurred in HIV-uninfected individuals.HTLV-1 infection was identified in 7 subjects with cryptococcal disease.Unfortunately, we did not know how many additional patients with HTLV-1 were tested.However, 7 of 38 cases represent a high percentage (18.4%) of coinfection between HTLV-1 and cryptococcosis, despite not having conducted additional testing of the cases.
The clinical and epidemiological characteristics of the Peruvian subjects and of those identified in the literature are shown in Table 1 [11-23, 25, 26, 28, 33-37].
Five of 7 Peruvian patients came from Andean and jungle regions.Mean age was 59 ± 8.9 years; 5 subjects were male and most of them had 1 comorbidity.Cryptococcus neoformans was the etiologic agent isolated in all subjects.
Meningeal cryptococcosis was detected in 6 of the 7 patients (85.7%).In the remaining patient, a lumbar puncture study was not performed, preventing the identification of meningeal cryptococcosis.Among the 6 patients in whom we identified meningeal cryptococcosis, only 1 patient had disseminated disease with cryptococcemia, pulmonary, gastrointestinal, and meningeal involvement.A baseline cerebrospinal fluid (CSF) latex agglutination antigen titer >1:1024 was found in 50% (n = 3/6) of the subjects, who also had altered mental status.CSF features of the 6 patients disclosed elevated opening pressure in 83.3% (n = 5), with a mean value of 33 cm H 2 O; 50% (n = 3) had India ink positive test; mean glucose level was 13 mg/dL (range, 2-29 mg/dL); median protein level was 100 g/L (range, 14-300 g/L), and median CSF white blood cell count was 148 cells/μL (range, 11-410 cells/μL).CSF fungal burden was measured in 50% (n = 3/6) of the subjects; the mean value was 7.39 ± 0.1 colony-forming units log 10 /mL and the mean isolation time was 5 ± 2.3 days.Lower fungal burden was observed in the remaining 50% (n = 3/6) of the subjects, with 7-10 days of growing time.
All 6 subjects received amphotericin B deoxycholate (0.6-1 mg/lg/day) plus fluconazole (800 mg/d) until obtaining a negative CSF culture.In 2 subjects, high opening pressure was maintained, and the CSF cultures remained positive after 21-40 days of treatment; both subjects died with Acinetobacter spp bacteremia.
Twenty-one publications with information on 29 patients with HTLV-1 and cryptococcosis disease were available for review.Most publications (12 [41.38%]) included Asian people.Data description was insufficient in 5 publications.The treatment was not standard in all cases.Amphotericin (AmB), fluconazole, 5-flucytosine (5-FC), miconazole, AmB plus 5-FC, fluconazole plus 5-FC, and lung surgery were prescribed.Survival information was available for 16 patients, with 7 of them remaining alive.

DISCUSSION
We report here 7 HIV-uninfected patients coinfected with HTLV-1 and cryptococcosis.In contrast to what has been reported from Asia and the Caribbean on this coinfection, where patients had essentially pulmonary involvement, our patients presented with meningeal involvement.However, more disseminated disease was not regularly investigated in all these studies including ours.Interestingly, patients coinfected with HTLV-1 and cryptococcosis received different treatment regimens.
Peru is an endemic country for HTLV-1, which has been reported in 2.5%-10% of Quechua ethnic groups from the mountains of Peru [38,39] and in 1.9%-5.9% of groups of native communities from the jungle region [40,41].Most of the 7 cases of this report were from the Andean and jungle regions of Peru.The precise etiology behind the elevated prevalence of HTLV-1 infection among Andean populations remains uncertain; certain researchers have hypothesized that factors such as geography, culture, or human leukocyte antigen types may contribute to this phenomenon [42].
HTLV-1 infection may increase susceptibility to opportunistic infections due to impaired immune response caused by regulatory protein expression.P30, a regulatory protein, reduces transcription in Toll-like receptor 4, leading to the inhibition of proinflammatory cytokines like monocyte chemoattractant protein 1, tumor necrosis factor α, and interleukin 8 [43].This suggests a potential mechanism for the heightened vulnerability to various pathogens in individuals with HTLV-1 infection.
The report highlights meningeal involvement in HTLV-1 and cryptococcosis cases, with specific characteristics such as pleocytosis, low glucose, and high protein concentration.India ink test results were negative in 50% of cases, and the fungal burden was notably high.In HIV-associated cryptococcal meningitis, the CSF white cell count is usually low and may even be normal.In contrast, like in other case series in patients with HTLV-1, we found pleocytosis in the CSF [13,22,26,34].This finding needs to be corroborated in larger studies.Disseminated cryptococcosis was not commonly investigated in this report or in others.CSF fungal burden data have not been extensively studied.In PWH, a fungal burden >4.5 colony-forming units log 10 /mL is linked to early clearance failure, but its correlation with antigen titers is unclear [44,45].Further research is needed to understand these associations fully.
Mortality is higher in HIV-uninfected and nontransplant patients compared to PWH and transplant recipients [46,47].In our description, 3 of 7 patients died.The mortality of previous reports was 56%.
In conclusion, we report 7 patients with HTLV-1 and cryptococcal infection.We recommend that patients with cryptococcosis should be tested for HTLV-1 infection in endemic areas for this human retrovirus.

Table 1 . Clinical Features, Treatment, and Outcome of Human Immunodeficiency Virus-Uninfected and Nontransplant Subjects With Human T-Lymphotropic Virus Type 1 and Cryptococcal Disease
BRIEF REPORT • OFID • 3