ABSTRACT
Chimeric antigen receptor (CAR) T cell immunotherapy has been a major advancement in cancer therapeutics. Reprogramming of T cells is achieved by using gammaretroviral or lentiviral vectors, which may interfere with human immunodeficiency virus type 1 (HIV-1) nucleic acid amplification testing (NAAT). Here, we describe three clinical scenarios in which CAR T cell immunotherapy interfered with HIV-1 testing, including (i) routine infectious disease screening prior to stem cell transplantation in a 16-year-old female with B cell acute lymphoblastic leukemia, post CAR T cell treatment; (ii) routine infectious disease screening prior to second CAR T cell collection in a 65-year-old male with diffuse large B cell lymphoma who failed initial CAR T cell treatment; and (iii) routine infectious risk assessment following an occupational health exposure from a 58-year-old male with multiple myeloma, who received CAR T cell treatment. In each case, patients initially tested negative by the “fourth-generation” HIV-1 screening enzyme immunoassay (targeting the p24 antigen and anti-HIV-1 antibodies), but positive by the Roche Cobas AmpliPrep/Cobas TaqMan HIV-1 test v2.0 (targeting gag and the long terminal repeat [LTR]). These samples subsequently retested negative using the Abbott m2000 RealTime HIV-1 assay, which targets the integrase gene. These results indicated that cross-reactions between lentiviral vectors and LTR genomes targeted in the HIV-1 NAAT caused the HIV-1 NAAT false-positive results.
INTRODUCTION
Chimeric antigen receptor (CAR) T cell therapy has become more accessible in clinical practice after the FDA approval of commercial products, and the use of CAR T cells as a treatment option for cancer is likely to continue to increase. Although this cutting-edge technology has helped aid in the treatment of cancer, one obstacle that has been reported for patients who have undergone this particular type of gene therapy is false-positive HIV-1 nucleic acid amplification testing (NAAT). CAR T cell therapy begins with mononuclear cell (MNC) collection, T cell purification, and reprogramming of the T cells to recognize receptors expressed by cancer cells, thereby targeting and killing the diseased cells. T cells are commonly reprogramed using HIV-1-derived lentiviral or gammaretroviral vectors (1). These vectors exploit HIV-1 genes encoding the viral envelope (env), GAG polyprotein (gag), and the polymerase (pol) to guide the CAR gene to the patient’s T cells, allowing it to integrate and transform the cells. Because of their more predictable integration profiles, lentiviral vectors are more commonly used for gene therapy (1, 2).
Cases of false-positive HIV-1 nucleic acid testing have been reported in the literature (3–5). These cases often were discovered during routine infectious disease testing as part of evaluations when further treatment such as stem cell transplantation (SCT) or additional CAR T cell infusions were necessary (6). Here, we describe additional clinical scenarios in which HIV-1 testing was clinically indicated and resulted in false-positive HIV-1 NAAT results, further underscoring the need for education of clinicians and laboratory staff on this potential pitfall and the implementation of appropriate assays to ensure timely and accurate HIV-1 testing for this patient population.
CASE REPORTS
Case 1.A 16-year-old girl with precursor B cell acute lymphoblastic leukemia (pre-B-ALL) received an investigational CD22 CAR T cell treatment at an outside institution in June 2018. Her routine infectious disease screening, which included HIV-1 serology and NAAT prior to MNC collection, was negative. The patient achieved remission post CAR T cell infusion and was admitted to our institution for bone marrow transplantation (BMT) from an HLA-matched sibling donor. Routine infectious disease screening, which included HIV-1 serology and NAAT, was performed prior to BMT. The initial screen revealed a nonreactive HIV-1 serology using the Architect HIV Ag/Ab Combo assay (Abbott Laboratories, Abbot Park, IL). However, HIV-1 NAAT performed using the Cobas AmpliPrep/Cobas TaqMan HIV-1 test v2.0 (CAP/CTM; Roche Molecular Systems, Inc., Pleasanton, CA) was positive, with a viral load of 20 copies/ml. The limit of detection for CAP/CTM is 20 to 10,000,000 copies (cp)/ml. At the time of testing, the patient was 35 days post CAR T cell infusion. Repeat HIV-1 NAAT performed at 37, 38, and 46 days post CAR T cell infusion on the same platform revealed slightly increased viral loads of 78, 171, and 167 copies/ml, respectively. The final sample collected at 46 days postinfusion was also tested using the m2000 RealTime system (Abbott Molecular, Des Plaines, IL), which revealed no viral RNA detected. The limit of detection for the Abbott m2000 is 40 to 10,000,000 cp/ml. Considering the patient’s age, her minimal risk for acute HIV-1, and the negative HIV-1 NAAT result using the Abbott m2000 platform, HIV-1 infection diagnosis was ruled out, and the patient proceeded to BMT. She achieved remission post BMT but relapsed 4 months later. Another round of CD19 CAR T cell treatment was pursued. Her routine infectious disease screening, which included HIV-1 serology (Architect) and quantitative HIV-1 PCR (CAP/CTM) prior to the second MNC collection, was negative.
Case 2.A 65-year-old gentleman with diffuse large B cell lymphoma (DLBCL) underwent MNC collection followed by infusion of an investigational CAR T cell product through a clinical trial in February 2018. His routine infectious disease screening, which included HIV-1 serology (Architect) and HIV-1 NAAT performed on the MNC product using Procleix Ultrio Plus HIV-1 assay (PUA; Gen-Probe, San Diego, CA), was nonreactive. The patient did not achieve remission from the initial CAR T cell infusion, and a second round of CAR T cell treatment was planned. Another MNC collection from the patient was performed in March 2018 for the manufacture of CAR T cell. He remained negative on the routine HIV-1 serology test prior to the second MNC collection. However, HIV-1 NAAT on the collection product using PUA was positive, and retesting of the patient using CAP/CTM was also positive, with a viral load of 320 copies/ml.
Additionally, the patient received 7 units of blood products (5 units of red blood cells and 2 units of platelets) at our institution after the CAR T cells were manufactured from his first MNC collection. To rule out transfusion-transmitted HIV-1 infection, the blood supplier was contacted, and all related donors were cleared by repeated HIV-1 testing (anti-HIV-1 antibody testing and HIV-1 NAAT), either of the donor or the donated product. Transfusion-transmitted HIV-1 infection was ruled out. Thus, the most likely etiology of false-positive HIV-1 test results remains cross-reactivity from CAR T cell infusion. Due to the rapid progression of his disease leading to death, it was not feasible to repeat HIV-1 testing in this patient.
Case 3.In July 2018, a 58-year-old male patient with multiple myeloma underwent MNC collection and CAR T cell infusion. Fifteen days later, a health care worker at our institution was splashed in the eyes with the patient’s bloody body fluid while working with him. Prior to CAR T cell infusion, his HIV-1 serology (Architect) was nonreactive and NAAT (CAP/CTM) was negative. To assess the risk of exposure from the source patient, HIV-1 serology by “fourth-generation” assay (Architect) and, although it was not a part of standard postexposure protocol, NAAT (CAP/CTM), were also performed. Serology was nonreactive; however, NAAT was positive, with a viral load of 4,091 copies/ml. Three days later (day 17 postinfusion), the patient had repeat positive NAAT (CAP/CTM) results with a viral load of 419 copies/ml. The latter sample was sent to an outside institution for repeat testing utilizing the Abbott m2000, and viral RNA was not detected by this method. The source patient was determined to be HIV-1 negative due to the repeated negative NAAT performed by the outside laboratory 1 week after the exposure occurred.
DISCUSSION
With increasing availability of new medical technologies, medicine is moving toward a more personalized approach. CAR T cells are a great example of therapy that allows a tailored treatment by reprograming a patient’s own T cells to help fight cancer. Currently, there are two FDA-approved products, Yescarta (axicabtagene ciloleucel), approved to treat diffuse large B cell lymphoma, and Kimyrah (tisagenleucel), approved for the treatment of leukemia. These treatments show great promise in the treatment of cancer, and new clinical trials are ongoing to develop additional CAR T cell gene therapy products to target other types of cancers, such as leukemia, lymphoma, lung, breast, and ovarian cancers, in both adult and pediatric populations, as well as other genetic disorders, such as X-linked severe combined immunodeficiency (X-SCID). With the broader availability of this form of therapy there is increased recognition of the interference of the CAR T cell products with HIV-1 NAAT testing in these patients. The first case report in 2014 described patients enrolled in a clinical trial for the use of CAR T cell therapy to treat X-SCID, a genetic disorder (3). Investigators found that two patients who were tested for HIV-1 as part of routine follow-up evaluation at 2 and 6 months posttreatment tested positive by HIV-1 NAAT. These test results were determined to be false positive by performing viral coculture assays to access for infectious replicating virus. Similarly, in 2017 and 2018, two additional institutions reported false-positive HIV-1 NAAT results in patients who received investigational CD19 CAR T cells for treatment of leukemia and lymphoma (4, 5). In these cases, HIV-1 testing was performed as part of routine screening prior to hematopoietic stem cell transplant (HSCT; Table 1). HIV-1 viral loads in these patients were, on average, below 500 copies/ml, except for that of one X-SCID patient who was found to have a viral load of almost 20,000 copies/ml 1 year postinfusion. The high viral load was related to T cells harboring portions of the lentiviral vector and undergoing expansion. Here, we report three cases in which interpretation of HIV-1 NAAT testing was complicated by CAR T cell therapy.
Summary of reported patients with false-positive HIV-1 NAAT test results following CAR T cell therapy
As CAR T cell therapy becomes more broadly accessible for cancer patients, clinicians should use caution when ordering HIV-1 NAAT testing for these patients. If HIV-1 NAAT testing is required for evaluation, the testing platforms must be considered. For all reported patients with false-positive HIV-1 results, testing had been performed by CAP/CTM. This assay, which targets the gag gene and long terminal repeat (LTR) of HIV-1, cross-reacts with the engineered T cells, which harbor lentiviral vectors that encode portions of HIV-1 homologous genes, causing the false-positive reaction. Thus, an alternative platform should be used for HIV-1 testing in CAR T cell patients. The Abbott m2000 and Versant HIV-1 RNA 3.0 (Siemens Diagnostics) assays, which target the integrase of the HIV-1 pol gene, are potential alternatives. In published cases, the Abbott m2000 assay was ultimately used to determine whether patients were truly negative. Other platforms that could potentially cause false-positive HIV-1 NAAT results in CAR T cell patients include the Aptima HIV-1 Quant assay (Aptima; Hologic, Inc, Marlborough, MA), which targets the pol gene and LTR. If information is available, the lentiviral vector utilized for the development of the CAR T cell product should be considered to facilitate optimal selection of HIV-1 NAAT platforms. Any known platforms that target the LTR, gag, or pol for the detection of HIV-1 should not be used because of the potential for cross-reaction with the lentiviral vectors.
Here, two of the three cases we report represent additional unique situations in which false-positive HIV-1 NAAT results could be encountered among CAR T cell patients. This includes testing of CAR T cell patients who have also received blood products. Since the 1990s, the FDA has required that all blood donor products be tested for HIV-1. The development of fourth-generation testing, which in addition to antibody also detects the p24 antigen, has enabled detection in patients that acquired the virus within 2 weeks of testing. The advent of HIV-1 NAAT testing further closed the window of detection by providing the ability to detect people who had acquired the virus within 2 days, thereby reducing the risk of transmission to 1 in 1,000,000 per unit of blood (7). Currently, there are 6 serological HIV-1 assays and 3 HIV-1 NAAT assays approved by the FDA for screening of blood donor products (https://www.fda.gov/vaccines-blood-biologics/complete-list-donor-screening-assays-infectious-agents-and-hiv-diagnostic-assays). Here, we describe a CAR T cell patient with false-positive HIV-1 NAAT who received multiple blood products between the time of his initial negative screen and the false-positive NAAT. This case was the first time our institution had encountered a positive NAAT result post CAR T cell infusion. This required identifying all donors to determine their HIV-1 status. Donors who made subsequent HIV-1-negative donations were not retested. This investigation not only took days to complete but delayed the time in which the patient could have apheresis of mononuclear cells to be prepared for a second CAR T cell infusion. This case highlights the importance of clinicians and laboratorians in the selection of the test platform used in the screening of CAR T cell patients.
Another consideration for HIV-1 testing of CAR T cell patients is in the event of an occupational exposure. According to the CDC, the risk of HIV-1 transmission due to an occupational exposure is 1 in 2 million, and there have 58 cases of reported and confirmed postexposure transmission of HIV-1 in the United States since the late 1990s (7). The U.S. Occupational Health and Safety Administration (OSHA) has specific guidelines for bloodborne pathogens (standard 29 CFR 1910.1030) that require employers to identify, obtain consent from the source patient, and test the source patient in a rapid manner (https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.1030). Additionally, the health care worker who has been exposed must be evaluated, during which postexposure prophylaxis may be considered if the risk is deemed necessary (8, 9). When investigating an occupational exposure, a rapid HIV-1 test must be utilized to appropriately evaluate transmission risk and may determine whether postexposure prophylaxis is necessary (8).
Testing of source patients usually includes a fourth-generation HIV-1 assay such as the Architect. For assessing occupation risk exposure, our institution utilizes the Architect assay. HIV NAAT is not currently a part of postexposure screening at our institution. In this case presentation, although the CAR T cell source patient was tested by the Architect assay, HIV-1 NAAT was also performed. Our institutional employee health and wellness policy indicates that the source patient should be tested for HIV; however, the policy does not specify what test should be used. When evaluating a source patient who has had CAR T cell therapy, clinicians and laboratorians must be aware of both current postexposure testing guidelines and, if NAAT testing is chosen, testing platform should also be considered for this patient population. This case presentation demonstrates that, if not initially recognized, testing CAR T cell patients using an inappropriate HIV-1 NAAT platform will not only result in a false-positive result but also prevent timely results for exposed health care personnel, as well as leading to emotional consequences for both the source patient and the potentially exposed personnel. Institutions should have appropriate protocols for HIV-1 testing of CAR T cells patients that are conducive to time-sensitive situations such as occupational exposure testing.
Here, we reviewed and presented case scenarios in which HIV-1 NAAT testing among CAR T cell patients produced false-positive results, highlighting the importance of careful assay selection and performance among this patient population. The consequences of false-positive HIV-1 NAAT results include not only time delays for downstream interventions, as well as for occupational exposure investigation, but also can become costly to not only the patient but also to the laboratory/institution because of continuous repeat testing necessary for investigating and determining whether results truly reflect a patient’s HIV-1 status. As this new therapy moves forward in becoming a standard of care for cancer treatment, clinicians and laboratory directors need to have a protocol in place that provides a list of assays appropriate for testing of patients that had CAR T cell therapy.
In conclusion, as CAR T cell therapy becomes more widely available for multiple indications, it is important for (i) clinicians to recognize the potential pitfall with routine HIV-1 NAAT and (ii) laboratories to have appropriate HIV-1 NAAT testing platforms readily available for this patient population, especially in time-sensitive situations such as occupational exposure testing.
FOOTNOTES
- Received 28 August 2019.
- Returned for modification 21 September 2019.
- Accepted 30 October 2019.
- Accepted manuscript posted online 6 November 2019.
- Copyright © 2019 American Society for Microbiology.
