High prevalence of plasma EBV DNA among the HIV positive individuals, with or without malignancies, attending the clinic at AIIMS, New Delhi

Abstract

Epstein-Barr Virus (EBV) is associated with the Hodgkin’s and Non-Hodgkin’s lymphoma (HL and NHL respectively). HIV is a risk factor for EBV infections and previously published data indicate that HIV infected individuals have higher chances of getting EBV infections compared to HIV uninfected individuals. Very limited information is available from India about the the prevalence of EBV in HIV positivity, with or without malignancy. In a recent study (Sinha et al. Current HIV Res 16:1–6, 2018) from All India Institute of Medical Sciences (AIIMS), New Delhi, we have shown that 2% among the HIV-1 infected individuals have malignancies including HL and NHL. To determine the prevalence of EBV among these individuals, clinical specimen obtained from ART clinic of AIIMS were tested for the presence of EBV DNA in plasma samples by quantitative real-time PCR. We have observed high prevalence of EBV (30%) among the 92 specimen tested. Prevalence is higher in patients with malignancy (37%) compared to those without maliganancy (27%). No correlation was observed with the CD4 counts or HIV viral load with EBV positivity.

HIV/AIDS is a risk factor for oncogenic virus infection such as Epstein-Barr virus (EBV), Human herpesvirus-8, Hepatitis B virus and Human papillomavirus [10, 12, 16]. Non-Hodgkin’s lymphoma (NHL) and Kaposi’s sarcoma are the major AIDS associated malignancies termed AIDS defining cancer (ADC), whereas Hodgkin’s lymphoma (HL), lung cancer etc. are the non-AIDS defining cancers (NADC) [6, 9]. In India, about 2.5 million people are currently living with HIV/AIDS. In recent times, the incidence of HIV infection is declining because of free access to anti-retroviral therapy (ART) through the ART clinics located in government hospitals [2, 5]. However, emergence of drug resistant strains of HIV, ART associated side effects and malignancies associated with AIDS continue to remain the major problems in the clinical management of AIDS patients. ADC and NADC are not uncommon among the AIDS patients in India. There are very limited studies done from India to determine prevalence of malignancies among AIDS patients. A large cross-sectional study was done recently from the All India Institute of Medical Sciences (AIIMS), New Delhi to estimate the prevalence and types of cancers among HIV infected individuals. 2% of the HIV infected individuals were found to have malignancies and cervical cancer (CA cervix) [13], NHL and HL were found to be the major types. EBV is associated with both HL and NHL. Therefore, we were interested to determine the prevalence of EBV infections among HIV infected individuals attending the ART clinic including those with proven malignancies.

We tested plasma samples from 92 HIV infected individuals among which 16 individuals had proven malignancies. Brief methods for the study are described here. DNA was isolated from 250 µL of plasma or serum using the QIAamp DNA blood mini kit (Qiagen Inc, Valencia, CA, USA) according to manufacturer instructions. A primer pair complimentary to the conserved BamH-W region of the EBV genome (the forward primer 5′-CCCAACACTCCACCACACC-3′ and the reverse primer 5′- TCTTAGGAGCTGTCCGAGGG-3′) was added (final primer concentrations = 300 nM) to DNA isolated from plasma. An oligonucleotide probe complimentary to a sequence within the BamH-W region of EBV nested within the region to be amplified (5′-(6-FAM)CACACACTACACACACCCAC CCGTCTC (BHQ-1)-3′) was added (final concentration = 25 nM) to PCR Supermix 2 × Reagent (Invitrogen, Carlsbad, CA) containing dNTPs and recombinant Taq DNA Polymerase (1 U/reaction). Reaction volumes were dispensed into 96-well PCR plates and Quantitative PCR was performed on a Bio-Rad iCycler iQ or MyiQ Real Time PCR Detection System (Bio-Rad, Hercules, CA). A calibration curve of known viral DNA copies (Namalwa cell line genomic DNA, ATCC #CRL-1432, A kind gift from Dr. Richard Ambinder) was run in parallel with each PCR analysis.

The plasma samples of 92 HIV-1 infected individuals who attended the AIIMS ART clinic including 16 with malignancies were tested for EBV DNA by quantitative real-time PCR. EBV genome was detected in 27 samples out of which 5 had EBV copies of more than 1000/ml, 12 had EBV copies of 100–1000/ml and 10 had copy numbers less than 100/ml (Table 1).

Table 1.

Distribution of EBV positivity among samples tested

	Sample screened	EBV PCR positive
Total	92	27  > 1000 copies = 5 100–1000 copies = 12  < 100 copies = 10
Male	62	22
Female	30	5
Without cancer	76	21 (27%)	p = 0.57
With cancer	16	6 (37%)

Among these 27 EBV positive individuals, no correlation was observed between the CD4 count or HIV viral load with the EBV copy number present in the plasma samples. 6 out of total 16 samples (37%) belonging to HIV positive individuals with cancer were positive for EBV DNA whereas 21 out of 76 samples belonging to HIV positive individuals without cancer (27%) were positive for EBV DNA, thus demonstrating a higher prevalence of EBV among HIV positive individuals with cancer as compared to those without cancer. We have done further analysis of this data and calculated p value was 0.567 (using chi-sqaure test), suggesting this data is not statistically significant.

Among HIV positive individuals with cancer and EBV PCR positivity, 4 out of 6 had HL (66%), 1 had NHL and one had gall bladder cancer (CA GB). Among EBV negative individuals, 4 had NHL, 1 had HL and 5 had CA cervix (Table 2). The mean CD4 count among the HIV positive EBV positive individuals with malignancy was 114/uL whereas among HIV positive EBV negative individuals with malignancy, the mean CD4 count was 222/uL.

Table 2.

Types of cancers among EBV positive and EBV negative HIV infected individuals

Group	Cancer type	Average baseline CD4
EBV positive	HL = 4 NHL = 1 CA GB = 1	114
EBV negative	HL = 1 NHL = 4 CA Cervix = 5	222

HIV infection is associated with increased B cell activation, which is known to induce EBV reactivation and EBV shedding [3, 8]. Previous studies have shown that the prevalence of EBV among HIV infected individuals is higher as compared to HIV uninfected individuals. Detection of EBV is done either by serology or genome detection by PCR. A few studies previously published from India have shown a high seroprevalence of EBV in both HIV infected and HIV uninfected individuals. However, none of these studies had done plasma EBV genome detection by PCR [11]. In an earlier cross-sectional study, we had recently demonstrated that the prevalence of malignancy among HIV infected individuals is 2%. In that study HL, NHL and cervcal cancer were found to be the most frequent types. This study was designed to determine the prevalence of EBV genome among both HIV positive individuals with as well as without cancer. Aim of the study was to investigate if HIV-1 positive individuals from this geographical region are at risk of EBV associated malgnancies or not. It has beev recently demonstrated that presence of plasma EBV DNA is better marker for EBV associated malignancy than EBV DNA in PBMC or saliva [4]. In the present study we have found that 30% of all subjects included were EBV positive which is in agreement with previous results. Several recent studies however shown that EBV DNA positivity ranging from 80 to 100% among HIV-1 positive individuals which are much higher than observed in the present study [1, 7, 11]. One explanation is that these studies included either total PBMC DNA or DNA extracted from saliva instead of cell-free plasma. Few systematic studies were previously conducted to compare the presence of EBV DNA in plasma and in PBMC and its correlation with EBV associated malignancies. Data from these studies have shown that the EBV DNA copies present in PBMCs are always higher than in plasma in both HIV + and HIV negative individuals without malignancy but in individuals with EBV associated malignancy the EBV DNA copies in both PBMCs and plasma are comparable [4, 14, 15]. As we expected, a higher percentage of HIV infected individuals with cancer tested positive for EBV genome, however this data did not show any statistical significance. EBV positivity was found to be associated with HL (75%) rather than NHL (18%). Our study is a small study with sample size of 92 out of which only 16 had cancer. More studies with larger sample sizes are recommended to better understand the dynamics of EBV infection among HIV positive individuals. No relationship was observed between the CD4 count and plasma EBV copy number among the HIV positive samples. However, there was a trend towards lower mean CD4 counts among the EBV positive HIV patients with cancer as compared to EBV negative HIV patients with cancer. A study with a larger sample size may be needed to demonstrate the significance of this observation. Our data suggests that EBV seroprevalence among HIV infected individuals is high in this geographical region of India which is similar to other published data. During AIDS progression, high levels of EBV may contribute to mortality by causing malignancies. Therefore, treatment of EBV infection alongside treatment of HIV with ART may result in better clinical outcomes.