The human T cell lymphotropic virus type (HTLV) was the first human retrovirus to be identified (1) and four main types are described: HTLV-1 to 4 (2–4), being HTLV-1 and HTLV-2 is the most prevalent. HTLV-1 is associated with an inflammatory neurological disease known as HTLV-1-associated myelopathy (HAM) (5) and a hematological malignancy known as adult T-cell leukemia/lymphoma (ATLL) (6). Furthermore, HTLV-1 has also been associated with several chronic inflammatory syndromes like uveitis, polymyositis, arthritis, alveolitis, infective dermatitis, some types of skin lesions and susceptibility to opportunistic infections (7–9). HTLV-1/2 is transmitted sexually, by sharing of intravenous equipment, by transfusion of cellular blood products, transplantation of organs or tissues, and from a mother to offspring in utero prior to labor, during delivery and through breastfeeding (10).

There are ~5–10 million HTLV-infected people worldwide (10). The endemic areas include Southwestern Japan, the Caribbean, Sub-Saharan Africa, South and Central America, Iran, and Melanesia (10). Brazil is considered one of the main endemic areas in South America, where it is estimated that at least 800,000 to 2.5 million people are infected with HTLV-1/2 (11). The HTLV-1 prevalence in Brazil varies according to geographical regions of the country (12), and the majority of cases are present in the urban population while HTLV-2 is more prevalent in indigenous tribes and among intravenous drug users (13, 14).

Since 1993, for transfusion purposes, screening of voluntary blood donors for HTLV-1/2 has been mandatory in Brazil (15). Despite this procedure and the high estimates of HTLV-infected individuals in the country, HTLV infection is still considered a neglected communicable disease since there is no compulsory notification to health authorities. Additionally, the scarcity of accurate diagnostic tests leads to the lack of a validated diagnostic algorithm. Accurate diagnostic tests are a powerful tool to control HTLV transmission.

In Brazil, the recommended algorithm for HTLV-1/2 diagnosis is based on conventional serological methods such as enzyme-linked immunosorbent assay (ELISA), particle agglutination, and chemiluminescence immunoassay. Western blot (WB) and/or line immunoassay (INNO-LIA) are usually applied as confirmatory and discriminatory diagnoses. However, indeterminate WB results can reach approximately 50% (16, 17), leading to an inconclusive diagnosis. In addition, the use of WB and line immune assay is not feasible for many developing countries, mainly due to the high cost. Moreover, ELISA positive or undetermined sera by ELISA were not always confirmed neither by WB nor line immune assays in Brazil.

The use of nucleic acid testing (NAT) platforms for HTLV viral gene detection instead of WB or line immune assay can be applied due to the higher sensitivity and specificity. Several centers have developed their own NAT assays generically called in-house tests (18–20). However, there is difficulty in controlling the quality of these tests, particularly because each laboratory uses distinct NAT strategies, such as genome targets, detection methods, and incomplete and/or inadequate methodological validation processes (20–22). These aspects make it challenging to standardize proviral detection limits, as well as to quantify proviral load, which is also an essential parameter for monitoring HTLV-infected individuals. Finally, these studies also showed a low number of samples tested by PCR compared to our study, which restricts the implementation of these molecular tests as routine or confirmatory HTLV diagnosis (18–20). One of the few studies that validated a real time PCR for intrapatient proviral load variability showed that the molecular test is suitable for HTLV diagnosis (23). Additionally there is no confirmatory nucleic acid (NAT) test for HTLV diagnosis which is approved by the Brazilian sanitary.

Thus, we have developed and validated a molecular platform with high sensitivity using a qualitative multiplex real-time PCR (duoplex-RT-PCR-HTLV-1/2) for both screening and discriminatory diagnosis of HTLV-1 and 2 infections. Additionally, this platform has been validated as a multicentric study in four main HTLV centers in Brazil, located in the states of São Paulo (two centers), Bahia, and Pará. The last two are where the majority of HTLV positive cases in Brazil are concentrated.

In the current study, we optimized and validated a qualitative multiplex qPCR platform—duoplex-RT-PCR-HTLV1/2—for the simultaneous detection and discrimination of HTLV-1 and HTLV-2. The established assay is sensitive and reproducible, easy to perform in molecular biology laboratories with appropriate conditions for clinical diagnosis; it is essentially a low-cost operation compared to the available confirmatory methods such as WB and line immunoassay.

Human T cell lymphotropic virus molecular diagnosis is a significant challenge in regard to standardization, appropriate verification, and validation. Several PCR methods have been reported for HTLV-1/2 diagnosis, but the complete validation procedure was not achieved (18, 25–27). Herein, we followed the main validation steps including analytical and diagnostic sensitivity, analytical and diagnostic specificity, robustness, and precision. Our duoplex-RT-PCR-HTLV1/2 platform demonstrated 94.6% of diagnostic sensitivity for HTLV-1 and 78.6% for HTLV-2 detection. Other studies that establish quantitative PCR demonstrate variable sensitivity. For example, Estes and Sevall (25) demonstrated sensitivity of 81.2% and 87.5% for HTLV-1 and for HTLV-2, respectively. Waters et al. (18) described a sensitivity of 100% for HTLV-1 and 62% for HTLV-2. In our work, for HTLV-1 detection, high diagnostic sensitivity has been observed, while the achieved sensitivity for HTLV-2 diagnosis remains much lower. The lower sensitivity for HTLV-2 in our assay might be related to several reasons, despite the effort of an additional accessory designed reverse primer (HTLV-2 Rev2_pol, see Table 1). The possible reason might be that the HTLV-2 proviral load, in general, is lower compared to HTLV-1 thus leading to lower sensitivity in viral detection (28). Moreover, HTLV-2 can also represent interspecies genetic divergence with the predominance of a specific subtype in our tested population, which can influence the performance of the diagnostic test. The analytic sensitivity for HTLV-2 was higher compared to HTLV-1 (10.9 vs. 3.9 copies/reaction). To address if HTLV-2/HIV-1 infection may affect the test sensibility, positive samples for both viruses were evaluated, and only one of eight samples was missed by the qualitative duoplex qPCR-HTLV-1/2 platform. We believe that neither immune suppression nor antiretroviral treatment for HIV-1 infection was related to this negative result. Similarly, Estes and Sevall noted similar sensibility to HTLV-2 (80%) using a multiplex approach (25). Our qualitative duoplex-RT-PCR-HTLV1/2 assay did not present amplification in all HTLV-1/2 non-infected samples, and thus reached 100% diagnostic specificity. Andrade et al. (29) demonstrated 98.5% of diagnostic specificity indicating the presence of false positivity. In fact, there are several qPCR methodologies for HTLV detection; some of them use SYBR^® Green, Molecular Beacons, and TaqMan^® probes (30). As mentioned before, these techniques have been developed to detect the most conserved regions of the HTLV 1/2 genome, including genes like gag, tax and pol (30). The use of different methodologies and target regions could impact directly on the sensitivity of the diagnostic test.

For the multicentric evaluation, we included four of the main clinical HTLV centers in Brazil with primary attendance and follow-up of HTLV-positive individuals. The optimized kit and training program were also provided in each location. The samples tested in this step were previously confirmed as HTLV-positive by another method such as nested-PCR or real-time PCR. As shown in Table 3, we achieved 91.1% of concordance between the tested samples. The discordant cases may be related to differences in the sensitivity of the HTLV PCR diagnosis adopted by each participating center. The highest concordance for HTLV-1 was achieved with Bahia and Para states, 96.5% and 89.3%, respectively, where also real-time PCR was used. In São Paulo city, discordant results were obtained in 20.6% of the cases. Nevertheless in this location the HTLV confirmatory diagnosis is based only on conventional nested-PCR. We also highlight that the majority of discordant cases were related to the presence of HTLV-2, where our assay showed lower detection sensitivity (78.6%) compared to HTLV-1.

Up to now, there is no molecular confirmatory platform for HTLV detection available commercially. The development and the detailed validation of our duoplex qPCR technology for confirmatory and discriminatory diagnosis of HTLV-1/2 infection will allow for the confirmation of the positive serological results and can offer a final conclusion to the obtained inconclusive results by EIA or WB. The established multiplex real-time PCR based molecular platform is currently under Brazilian (BR 10.2014.024905.2) and international patent applications (United States of America No. 15/517.449 and European Union patent application No. 158484220).

In conclusion, we developed, optimized and validated a duoplex qPCR qualitative test for HTLV-1/2 diagnosis (confirmatory and discriminatory), which is ready-to-use, timesaving and presents low risk for sample cross-contamination. This technique was performed in a unique reaction tube to detect and discriminate between HTLV-1 and HTLV-2. Performance tests showed high sensitivity and specificity compared to the singleplex assays and to classical confirmatory methods for HTLV-1/2 diagnosis such as WB. Therefore, this platform can be implemented as a fast HTLV diagnosis and confirmation and to monitor HTLV-infected patients in the clinical centers for the correct management as well as counseling to avoid transmission. Despite the obtained suitable validation parameters of our assay, we highlight that it is important to include more HTLV-2-infected patients and perform the validation processes in other national and international centers engaged in HTLV diagnosis.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Clinical Hospital of School Medicine—Process number 20009119.0.1001.5440. The Ethics Committee waived the requirement of written informed consent for participation.

MR-J idealized the proposal, performed major experiments and analysis, and wrote the original draft. ER, SS, TA, MP, and DL performed major experiments and analysis and reviewed the original draft. MS supervised sample collection and diagnosis of HTLV-1-infected and healthy individuals. VO performed major experiments and analysis and reviewed the original draft. BG-C, BF, AP, JS, OT, and MS supervised sample collection and diagnosis of HTLV-1-infected and healthy individuals, as well as ambulatory follow-up of HTLV-1 infected individuals. JC, DC, and SK idealized the proposal, supervised and reviewed the original draft. All authors contributed to manuscript revision, read, and approved the submitted version.

This work was supported by Financiadora de Estudos e Projetos (FINEP/Process number 1387/10), Centro de Terapia Celular (CTC/FAPESP)/Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP—Process number 2013/08135-2), FAPESP n. 2018/07239-2; Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq—Processo 465539/2014-9), CNPq n. 301275/2019-0, CNPq n. 301677/2018-2, Fundação Hemocentro de Ribeirão Preto (FUNDHERP); and scholarship from FM/FMUSP to JC.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.