Field Study of Dried Blood Spot Specimens for HIV-1 Drug Resistance Genotyping C. M. Parry,a,b N. Parkin,c K. Diallo,d S. Mwebaza,e R. Batamwita,e J. DeVos,d N. Bbosa,f F. Lyagoba,a B. Magambo,a M. R. Jordan,g R. Downing,f G. Zhang,d P. Kaleebu,a C. Yang,d S. Bertagnolioh

Dried blood spots (DBS) are an alternative specimen type for HIV drug resistance genotyping in resource-limited settings. Data relating to the impact of DBS storage and shipment conditions on genotyping efficiency under field conditions are limited. We compared the genotyping efficiencies and resistance profiles of DBS stored and shipped at different temperatures to those of plasma specimens collected in parallel from patients receiving antiretroviral therapy in Uganda. Plasma and four DBS cards from anti-coagulated venous blood and a fifth card from finger-prick blood were prepared from 103 HIV patients with a median viral load (VL) of 57,062 copies/ml (range, 1,081 to 2,964,191). DBS were stored at ambient temperature for 2 or 4 weeks or frozen at ⴚ80°C and shipped from Uganda to the United States at ambient temperature or frozen on dry ice for genotyping using a broadly sensitive in-house method. Plasma (97.1%) and DBS (98.1%) stored and shipped frozen had similar genotyping efficiencies. DBS stored frozen (97.1%) or at ambient temperature for 2 weeks (93.2%) and shipped at ambient temperature also had similar genotyping efficiencies. Genotyping efficiency was reduced for DBS stored at ambient temperature for 4 weeks (89.3%, P ⴝ 0.03) or prepared from finger-prick blood and stored at ambient temperature for 2 weeks (77.7%, P < 0.001) compared to DBS prepared from venous blood and handled similarly. Resistance profiles were similar between plasma and DBS specimens. This report delineates the optimal DBS collection, storage, and shipping conditions and opens a new avenue for cost-saving ambient-temperature DBS specimen shipments for HIV drug resistance (HIVDR) surveillances in resource-limited settings.

M

ost people affected by the HIV/AIDS pandemic live in resource-limited settings. Tremendous progress has been made in the scale-up of antiretroviral therapy (ART) in resourcelimited settings (1); however, increased access to ART raises concern about the emergence and transmission of significant population-level HIV drug resistance (HIVDR) (2–5). In resource-rich countries, HIV patient monitoring includes routine viral load (VL) testing and HIVDR genotyping prior to treatment initiation and in the setting of virological failure (6). In most resource-limited settings, routine VL monitoring and HIVDR testing are not widely available or economically feasible with current technologies. To address concerns about emergence and transmission of HIVDR, the World Health Organization (WHO) recommends a public health surveillance approach that includes monitoring the level of HIVDR in populations on ART (acquired HIVDR), in patients eligible for ART (pretreatment HIVDR), and in recently HIV-infected populations (transmitted HIVDR) (7, 8). Plasma specimens obtained from whole blood are considered to be the gold standard specimen type for HIVDR genotyping because HIV sequence data obtained from plasma are derived from actively replicating viruses that form the majority of viruses circulating in the body. Serum, including dried serum spots, has also been successfully used for HIVDR testing (9). Despite being the gold standard, the collection, preparation, storage, and transportation of plasma or serum specimens require laboratory equipment such as centrifuges and cold-chain facilities, including freezers and dry ice. Such facilities are not reliably available in many resource-limited settings, especially in more remote, rural areas. Omitting HIV-infected individuals in remote areas when performing surveillance of HIVDR not only may significantly bias

2868

jcm.asm.org

Journal of Clinical Microbiology

national estimates, thus limiting the utility of results for public health and ART program planning, but also may lead to a failure to detect the accumulation of drug-resistant HIV variants in these areas. As a result, there is an urgent need for field-friendly methodologies for the preparation, storage, and transport of specimens in resource-limited settings suitable for WHO-recommended surveys of HIVDR (8). An alternative specimen type to plasma whose use is more practical in the field in resource-limited settings is dried blood spots (DBS). DBS require a smaller volume of blood than plasma or serum, and their usage minimizes the need for cold-chain storage and transportation requirements. In addition, DBS specimens, once completely dried, are considered noninfectious and nonhazardous (10), allowing them to be shipped at reduced cost since they can be transported at ambient temperature using a standard courier service or in the cargo hold of regular passenger airline services. The use of DBS for performing national-level surveillance of transmitted, pretreatment, or acquired HIVDR would be

Received 26 February 2014 Returned for modification 2 April 2014 Accepted 22 May 2014 Published ahead of print 28 May 2014 Editor: Y.-W. Tang Address correspondence to N. Parkin, [email protected], or C. Yang, [email protected]. C.M.P., N.P., and K.D. contributed equally to this article. Copyright © 2014, American Society for Microbiology. All Rights Reserved. doi:10.1128/JCM.00544-14

p. 2868 –2875

August 2014 Volume 52 Number 8

Downloaded from http://jcm.asm.org/ on March 8, 2015 by DAHLGREN MEDICAL LIBRARY

MRC/UVRI Uganda Research Unit on AIDS, Uganda Virus Research Institute, Entebbe, Ugandaa; MRC/UCL Centre for Medical Molecular Virology, UCL, London, United Kingdomb; Data First Consulting, Belmont, California, USAc; Division of Global HIV/AIDS, CGH, CDC, Atlanta, Georgia, USAd; Mildmay Centre, Kampala, Ugandae; CDCUganda, Entebbe, Ugandaf; Division of Geographic Medicine and Infectious Disease, Tufts University School of Medicine, Boston, Massachusetts, USAg; HIV Department, World Health Organization, Geneva, Switzerlandh

Field Study of DBS on HIV-1 Drug Resistance Genotyping

MATERIALS AND METHODS Study subjects, specimen preparation, storage, and shipment conditions. This study targeted HIV-infected patients with plasma VL of over 1,000 copies/ml who were on ART for at least 6 months and who provided written informed consent. Initially, during June and July 2011, patients visiting the Mildmay Centre in Kampala, Uganda, for regular ART drug pickup were screened for eligibility in the study. In the first 2 weeks of specimen collection, 63 patients were recruited. VL tests revealed that only two patients had VL of over 1,000 copies/ml and were thus eligible for inclusion in the study. To facilitate recruitment, patients suspected of having virological failure based on previous VL test results, CD4 cell counts, and/or clinical judgment were identified. Recruitment was continued between September and October 2011 at Mildmay and expanded to three additional clinics: the AIDS Support Organization (TASO) clinic in Entebbe and two Medical Research Council/Uganda Virus Research Institute (MRC/UVRI) clinics in Entebbe and rural southwest Uganda.

August 2014 Volume 52 Number 8

TABLE 1 Storage and shipping conditions of specimens Shipment condition

Group

Specimen type

Storage conditionb

1 2 3

Plasma DBSa DBSa DBSa

4 5

DBSa DBS (finger prick)

Immediately at ⫺80°C Immediately at ⫺80°C Immediately at ⫺80°C Ambient for 2 weeks and then ⫺80°C Ambient for 4 weeks Ambient for 2 weeks and then ⫺80°C

Dry ice Dry ice Ambient Ambient Ambient Ambient

a

DBS made from venous blood collected in a K2EDTA tube. All the DBS specimens prepared either from venous K2EDTA-anti-coagulated blood or directly spotting from finger-prick blood were allowed to dry overnight at ambient temperature before package and storage.

b

During the second phase of enrollment, 227 patients, 104 (45.8%) of whom had an HIV-1 load over 1,000 copies/ml, were identified. Overall, blood specimens were collected from 290 patients, 106 (36.6%) of whom had an HIV-1 load of over 1,000 copies/ml. Among the 106 patients with VL of over 1,000 copies/ml, 103 had a complete set of DBS from all groups and were included in this study. Venous blood was collected from each of the patients using a 10-ml K2EDTA tube by phlebotomy, and a DBS card was also prepared from the same patient by directly spotting from finger-prick blood (3 to 5 spots per patient). Blood tubes were transported to the MRC/UVRI laboratory in Entebbe, and DBS were prepared on the same day, while the finger-prick DBS were dried overnight on site and transported to the laboratory on the following day. Four DBS cards were prepared from each blood specimen using 75 ␮l/spot and 5 spots per card and dried overnight. The remaining blood was then centrifuged for plasma preparation. Plasma VL was determined using the Cobas Ampliprep/Cobas TaqMan (CAP/CTM) HIV-1 test, v2.0 (Roche, Basel, Switzerland). After drying, DBS were placed individually into glassine envelopes and packaged into tightly sealed ziplock bags containing desiccants and a humidity indicator card before being stored under different conditions as shown in Table 1. DBS cards were stored for 2 or 4 weeks at ambient temperature or frozen at ⫺80°C before being shipped using a standard international courier service at ambient temperature or frozen on dry ice as indicated in Table 1. During the study period, the average ambient temperature during storage was approximately 26°C and ranged from 19 to 32°C. Although humidity was not measured, the average relative humidity in the Entebbe region was 82%, ranging from 65% to 94% between June and December 2011. Plasma specimens were stored frozen at ⫺80°C, and at the end of specimen collection, an aliquot was shipped on dry ice to the Centers for Disease Control and Prevention (CDC) laboratory for HIVDR genotyping in Atlanta, GA. Upon receipt, the specimens were stored at ⫺80°C before testing. For all the ambient-temperature shipments, a Flashlink temperature data logger (DeltaTRAK, Pleasanton, CA) was packaged with the specimens to monitor temperature during the shipments and shipment routes were also determined from the courier service online tracking system. This study was approved by the Uganda Virus Research Institute Science and Ethics Committee and the Uganda National Council for Science and Technology. The genotyping of deidentified specimens conducted at the WHO-designated Specialized Drug Resistance Laboratory at CDC, Atlanta, GA, was deemed nonhuman-subject research by the Associate Director for Science at the Center for Global Health, CDC. HIVDR genotyping and sequence analysis. Genotyping of the protease and RT regions of the HIV-1 pol gene was carried out using a broadly sensitive in-house method (15, 31). Briefly, a 1,084-base-pair segment of the 5= region of the pol gene was generated by RT-PCR followed by nested PCR using total nucleic acid extracted from a DBS. This fragment was purified, sequenced using a BigDye Terminator v3.1 cycle sequencing kit (Life Technologies, Foster City, CA), and analyzed on an ABI Prism 3730

jcm.asm.org 2869

Downloaded from http://jcm.asm.org/ on March 8, 2015 by DAHLGREN MEDICAL LIBRARY

especially useful as DBS specimens can be collected from any individuals meeting the requirements of survey eligibility criteria in the field regardless of the availability of good laboratory infrastructure and cold-chain transportation systems. Studies evaluating the use of DBS for HIVDR testing have been carried out with various degrees of success and have been recently reviewed (11–13). Amplification of the protease and reverse transcriptase (RT) regions using a variety of testing methods has been reported. The sensitivities of PCR amplification from DBS differ significantly, with a few in-house assays reporting success from DBS specimens with VL as low as 1,000 to 2,000 copies/ml (14– 16). All reports indicate lower genotyping sensitivities from DBS than from plasma specimens (17–21). Notably, genotyping sensitivity of 1,000 copies/ml was achieved by only two laboratories in a recent WHO proficiency testing program (22). The contribution of cellular RNA or proviral DNA in DBS specimens has also been investigated, with one study documenting DNA contribution in over half of patients with undetectable VL and poor sequence reproducibility from replicate PCRs (23). In another study, proviral DNA contributed to 24 of the 37 sequences produced from DBS specimens; nonetheless, the nucleotide sequence identity of paired DBS and plasma specimens was high (98.1% to 98.8%) (24, 25). Based on these and other data, DBS specimens are recommended as a suitable substitute for plasma in HIVDR surveillance in resource-limited settings (14–16, 18, 21, 25–28). The efficiency of HIVDR genotyping from DBS appears to be dependent on the assay methodology and storage conditions (temperature and humidity). Studies have shown that when DBS were stored at higher temperature and humidity, the proportion successfully amplified and genotyped was reduced (19, 21, 27, 29, 30). Despite all these studies, there are limited data on the effect of shipment conditions on genotyping efficiency or whether previously frozen DBS can be shipped at ambient temperature using routine courier services without significantly compromising genotyping efficiency. A recent study performed in a controlled laboratory environment reported no significant impact of ambient-temperature shipment on genotyping efficiency in comparison to frozen shipment on dry ice (22). The aim of this study was to determine the optimal ambient-temperature storage and shipment conditions of DBS specimens for HIVDR genotyping using a relatively low-cost and broadly sensitive in-house genotyping assay (25). (These data were presented in part at the International Workshop on HIV & Hepatitis Virus Drug Resistance and Curative Strategies, 5 to 9 June 2012, Sitges, Spain.)

Parry et al.

Genetic Analyzer (Life Technologies). A customized RECall software program (32) was used to edit the raw sequence data and generate consensus sequences. The REGA HIV-1 subtyping tool, v2.0, was used for HIV-1 subtype determination (http://www.bioafrica.net/rega-genotype/html /subtypinghiv.html). Drug resistance mutations were identified using the Stanford HIVdb program (http://hivdb.stanford.edu/) (33). Nucleotide sequence analysis was performed using Microsoft Excel and MEGA 5.0 (34), and strict nucleotide identity was calculated (i.e., mixtures were counted as representing a difference if the base in the plasma-derived viral sequence was not a mixture). Statistical analysis. Differences in the genotyping efficiencies were analyzed using the Pearson chi-square test with the continuity correction or Fisher’s exact test where appropriate. To reduce the chance of making a type II error, no adjustments were made for multiple comparisons, as it was considered necessary to sensitively detect the differences in the genotyping efficiencies between the groups. VL values were log10 transformed before analysis. VL variation among shipments was assessed by one-way analysis of variance (ANOVA) and the least-significant-difference (LSD) post hoc test. Nucleotide sequence identity scores determined for plasma and different groups of DBS were compared using the Wilcoxon matchedpairs signed-rank test. P values were adjusted for false discovery using the Benjamini-Hochberg procedure (35). The data analyses were performed using SPSS 21.0 software (IBM SPSS Inc., Armonk, NY) and Prism 6.0 software (GraphPad Software Inc., La Jolla, CA). A 2-tailed P value of ⬍0.05 was considered statistically significant.

RESULTS

Patient clinical characteristics. The median VL of the 103 study participants was 57,062 and ranged from 1,081 to 2,964,191 copies/ml (interquartile range, 14,380 to 165,714). Eleven patients had VL between 1,000 and 5,000 copies/ml, and eight had VL between 5,000 and 10,000 copies/ml. The majority (56.9%, 58/102 [1 patient without data]) of the participants were on ART for over 3 years, ranging from 2 weeks to 10 years, and were receiving a standardized non-nucleoside reverse transcriptase inhibitor (NNRTI)-based first-line regimen, while 14.6% (15/103) were on a protease inhibitor (PI)-based second-line regimen. The following antiretroviral (ARV) drugs were included in the regimens: zidovudine (72%), lamivudine or emtricitabine (100%), tenofovir (29%), abacavir (4%), nevirapine (60%), efavirenz (20%),

2870

jcm.asm.org

lopinavir/ritonavir (14%), and indinavir/ritonavir (1%). The five DBS cards prepared from each eligible patient were stored and shipped under different study conditions as described in Materials and Methods (Table 1). Shipping conditions. The temperatures and durations of shipments between the MRC/UVRI facility in Entebbe, Uganda, and the CDC laboratory in Atlanta are shown in Fig. 1. Most shipments arrived at the laboratory after 4 days. Shipments were routed from Entebbe to Dubai, United Arab Emirates (Middle East and African hub for the international courier), and on to a hub in Europe (usually London, United Kingdom) before transatlantic flights, followed by short domestic flights to Atlanta. One shipment (number 6) included an additional leg from Dubai to Delhi, India, which added two more days to the total transport time (6 days to arrive at destination). The highest temperature recorded was 40°C, and lowest was 1°C; however, most shipment temperatures ranged between 35°C and 5°C. Drug resistance genotyping efficiency. The genotyping efficiencies (percentages of specimens for which amplification and sequencing were successful) of the different groups of specimens, grouped by VL range, are shown in Fig. 2. Using plasma, the overall proportion successfully genotyped was 97.1% (100/103); there was no significant difference among VL subgroups. The proportion successfully genotyped from DBS stored and shipped frozen (DBS group 1) was similar to the proportion successfully genotyped from plasma (98.1%, 101/103), which confirms the broadly sensitive nature of the in-house assay used (15). For group 2 DBS, which were stored immediately at ⫺80°C and shipped at ambient temperature, the overall proportion successfully genotyped was also high (97.1%, 100/103); an apparent reduction in the proportion of DBS with VL ⬍ 10,000 copies/ml (89.5%, 17/19) was not statistically significantly different from the plasma results (94.7%, 18/19). The overall proportion successfully genotyped for DBS stored at ambient temperature for 2 weeks before being stored frozen at ⫺80°C and shipped at ambient temperature (DBS group 3) was 93.2% (96/103), which is not statistically different from the results determined for the frozen DBS group (group 1) or plasma.

Journal of Clinical Microbiology

Downloaded from http://jcm.asm.org/ on March 8, 2015 by DAHLGREN MEDICAL LIBRARY

FIG 1 Temperature variations during ambient-temperature shipments. Temperatures were recorded using temperature data loggers included in the DBS shipping packages. The shipments were numbered sequentially (shipment 4 was omitted due to the missing of temperature data logger in the shipping package), and the numbers of patients whose DBS specimens were contained in each shipment are indicated in the legend.

Field Study of DBS on HIV-1 Drug Resistance Genotyping

Again, no difference was seen even when VL was below 10,000 copies/ml. However, DBS stored for 4 weeks and shipped at ambient temperature (DBS group 4) resulted in a statistically significant drop in the overall proportion successfully genotyped to 89.3% (92/103, P ⫽ 0.03 and 0.01 compared to plasma and DBS group 1, respectively). The reduction was mainly due to a large impact on the proportion of DBS successfully genotyped for specimens with VL ⬍ 10,000 copies/ml (63.2%, 12/19). The DBS specimens made from finger-prick blood that were stored at ambient temperature for 2 weeks before being frozen at ⫺80°C and shipped at ambient temperature (DBS group 5) had the lowest overall proportion successfully genotyped at 77.7% (80/103, P ⬍ 0.001 compared to plasma or DBS group 3). As for DBS group 4, the decrease was mainly due to the lower proportions successfully genotyped from DBS with lower VL levels: 47.4% at VL ⬍ 10,000 and 67.9% at VL between 10,000 and 50,000 copies/ml. Compared to plasma, the proportion successfully genotyped from DBS with VL ⬍ 10,000 copies/ml was reduced for all DBS groups except group 1 (stored immediately at ⫺80°C and shipped frozen). However, this effect was statistically significant only for DBS stored at ambient temperature for 4 weeks and shipped at ambient temperature (DBS group 4) and the DBS made from finger-prick blood (DBS group 5). While the temperatures during shipment were fairly consistent (Fig. 1), analysis of the shipment temperature data (only performed for shipments containing 10 or more specimens) suggested that the genotyping efficiency from DBS was the lowest in those shipments where the maximal temperatures reached over 30°C (i.e., shipments 2 and 4 compared to shipment 10; Table 2). However, the median VL in the specimens of shipment 10 was the highest of those evaluated, which could also have had an impact on the genotyping efficiency. Drug resistance mutation prevalence and sequence comparison. The presence of HIVDR-associated mutations was determined and is summarized in Table 3. Most patients (88.3%, 91/ 103) for whom an HIVDR genotype was obtained had mutations associated with nucleoside or nucleotide reverse transcriptase inhibitor (NRTI) resistance. The predominant NRTI resistance mu-

August 2014 Volume 52 Number 8

tations were M41L, D67N, K70R, L210W, T215F/Y, K219Q, and M184V. Similarly, most (87.4%, 90/103) patients also had mutations associated with resistance to NNRTI, predominantly, A98G, K103N, Y181C, and G190A. Only five (4.9%) patients had resistance to PIs even though 15 patients were being treated with a PI-containing regimen. The PI mutations identified were M46I, I54V, V82A, and L90M. Six patients (5.8%) had wild-type virus with no resistance mutations, four (3.9%) had only NNRTI mutations, and one (1%) had only PI mutations (V82V/L and L90L/ M). Subtyping analysis using the REGA HIV-1 subtyping tool, v2.0, revealed that 58.3% of sequences were subtype A1, 38.8% subtype D, and 2.9% subtype C. Pairwise sequence identity analyses of results from different DBS groups and plasma (considered the “gold standard”) indicated high overall sequence identity (mean pairwise identity over 98.9%) for both the entire sequence and just the codons at resistance-associated positions (Table 4). A small but statistically significant (P ⬍ 0.0001) decrease in sequence identity versus plasma was found with DBS shipped at ambient temperature (group 2; median, 99.43%) or stored at ambient temperature for any period of time (groups 3, 4, and 5, 99.09% to 99.24%) compared to DBS group 1 (99.52%; Table 4). When the analysis was restricted to nucleotide sequence identity in codons corresponding to posi-

TABLE 2 Viral load variation according to shipment Shipment no. n

Median log10 VL (IQRa)

Min Max Mean Shipment temp temp temp % duration (°C) (°C) (°C) genotyped

2 4 6 7 10

4.25 (3.82–4.69) 4.60 (4.35–4.85) 4.68 (4.36–4.99) 4.75 (4.29–5.21) 5.10 (4.66–5.54)

4 days 7 h No datab 6 days 0 h 4 days 6 h 4 days 5 h

15 28 14 14 11

15.3

32.5

25.7

1.0 12.7 7.8

26.9 30.7 29.6

20.3 20.7 19.7

91.1 94.0 92.9 92.9 100c

a

IQR, interquartile range. The data logger was omitted for this shipment. c P ⬍ 0.05 versus shipment 2 or 4 (one-way ANOVA with LSD multiple comparisons). For this analysis, only the 5 shipments with at least 10 specimens were included. b

jcm.asm.org 2871

Downloaded from http://jcm.asm.org/ on March 8, 2015 by DAHLGREN MEDICAL LIBRARY

FIG 2 Proportions of specimens successfully genotyped. The percentage of specimens in each group that were successfully genotyped is shown, grouped by viral load range. Plasma was frozen immediately after separation and stored and shipped frozen. See Table 1 for storage and shipment conditions for the different groups of DBS.

Parry et al.

TABLE 3 Drug resistance mutation prevalence % of specimens

Mutation

NRTI

Any M41L K65R D67G D67N T69N K70R L74I V75I V75M Y115F M184V L210W T215F T215Y K219E K219Q

91 35 8 5 26 10 26 6 6 8 6 88 25 20 25 8 18

88.3 34.0 7.8 4.9 25.2 9.7 25.2 5.8 5.8 7.8 5.8 85.4 24.3 19.4 24.3 7.8 17.5

NNRTI

Any A98G K101E K103N V108I Y181C Y188L G190A H221Y

90 17 9 43 11 21 5 29 11

87.4 16.5 8.7 41.7 10.7 20.4 4.9 28.2 10.7

PI

Any majorb

5

4.9

a

NRTI and NNRTI mutations detected in at least 5 plasma specimens are shown. b PI mutations observed: M46I (n ⫽ 2), G48M (1), I50V (1), I54V (3), V82A (3), V82L (1), L90M (3).

tions where resistance-associated mutations are known to occur, the median identity score was 100% for all groups (range in mean values from 99.18% to 99.36%), and no statistically significant decreases were observed (adjusted P ⬎ 0.05). Considering DBS specimens from individual patients, differences in the lists of resistance-associated mutations detected were occasionally observed between the groups. Table 5 shows DR mutation lists from two patients where low-resistance-codon sequence similarity was observed. In specimen A, mutations at positions 62, 69, 70, and 75 were detected in some but not all sequences; the presence of these mutations leads to predictions of higher-level resistance to several NRTIs than when they are absent. The mixture at RT position 179 (V179IT) was not detected in

DBS group 3 or 4, in a background of K103S, G190A, and P225H. The V179T mutation makes a small contribution to the predicted level of resistance to NNRTIs in the Stanford resistance algorithm. In sample B, the G190A mutation was reported to be a mixture (G190AG) in plasma and DBS groups 1 to 3 but to be unmixed in groups 4 and 5. In contrast, the M184V mutation, also present in plasma and groups 1 to 3 in a mixture, was not detected in group 4 or 5. T215Y was detected in the DBS prepared from finger-prick blood whereas T215F (group 4) or revertants (T215IST, groups 2 and 3) were detected in other DBS samples and were absent in plasma and DBS group 1. Again, these differences led to changes in the predicted level of resistance to several NRTIs. One explanation for these differences seen between DBS and plasma is that the proviral DNA present in DBS had mutation patterns different from those of free virus in the plasma. Alternatively, the differences could have been a result of inconsistent amplification of the mixtures of variants present at the PCR step. DISCUSSION

The results presented here demonstrate that storage of DBS at ambient temperature for 2 weeks, shipment of DBS at ambient temperature after being stored frozen, and up to 6 days of ambient-temperature shipment time do not significantly affect the proportion of specimens successfully genotyped compared to the frozen-plasma gold standard. Collection and long-term storage at ⫺80°C followed by shipment in batches to the genotyping laboratory are also acceptable. Our results also highlight conditions to avoid when using DBS for HIVDR genotyping, namely, storage at ambient temperature for longer than 2 weeks or use of finger-prick blood collection when VL is lower than about 50,000 copies/ml. Our study results indicate that use of DBS prepared from finger-prick blood leads to a reduction in genotyping efficiency. However, for those DBS with a VL of over 50,000 copies/ml, they could be still a suitable specimen type. The reasons for the reduction in genotyping efficiency, in spite of the relatively insignificant impact on VL testing (36), are unclear. Possible explanations include the reduced concentration of virus particles in capillary blood due to the dilution from tissue fluid present only in fingerprick blood, presence of interfering substances or nucleases, or lower absolute volume of blood spotted onto each spot from finger-prick blood compared to the precise spotting by pipetting anticoagulated venous blood. To our knowledge, this is the first study that monitored DBS specimens shipped in batches, over a period of weeks, between laboratories for HIVDR genotyping. As the packages were sent as nonhazardous material, they would have been included with other

TABLE 4 Nucleotide sequence and drug resistance mutation identities between paired DBS and plasma specimens % nucleotide sequence identity

% DR position identity

Group

n

Minimum

25% percentile

Median

75% percentile

Maximum

Mean

Minimum

25% percentile

Median

75% percentile

Maximum

Mean

1 2 3 4 5

99 98 94 90 78

95.8 97.33 96.08 96.85 93.7

98.95 98.95 98.57 98.64 98.47

99.52 99.43a 99.24a 99.14a 99.09a

99.81 99.71 99.62 99.62 99.52

100 100 100 100 100

99.31 99.18 99.06 99.04 98.96

91.23 94.74 89.47 91.23 94.74

98.25 98.25 98.25 98.25 98.25

100 100 100 100 100

100 100 100 100 100

100 100 100 100 100

99.36 99.25 99.24 99.18 99.3

a

Wilcoxon matched-pairs signed-rank test P ⬍ 0.0001 versus group 1.

2872

jcm.asm.org

Journal of Clinical Microbiology

Downloaded from http://jcm.asm.org/ on March 8, 2015 by DAHLGREN MEDICAL LIBRARY

Drug class

No. of specimensa

Field Study of DBS on HIV-1 Drug Resistance Genotyping

TABLE 5 Drug resistance mutations for selected patients with low drug resistance mutation profile similaritya Group

NRTI DRMs

NNRTI DRMs

% similarity to plasma at DR sites

A

0 1 2 3 4

A62AV, V75IV, M184V A62AV, V75IV, M184V M184V T69N, K70R, M184V A62AV, T69NT, K70KR, V75IV, M184V

K103S, V179AITV, G190A, P225H K103S, V179IT, G190A, P225H K103S, V179IT, G190A, P225H K103S, G190A, P225H K103S, G190A, P225H

Not applicable 100.0 96.5 93.0 94.7

B

0 1 2 3 4 5

M41LM, T69ADNT, M184MV M41LM, M184MV M41LM, T69NT, M184MV, T215FIST M41LM, M184MV, T215FIST D67N, T215F M41L, T69N, T215Y

V90I, K103N, G190AG V90I, K103N, G190AG V90I, K103N, G190AG V90IV, K103N, G190AG V90I, K103N, G190A V90I, K103N, G190A

Not applicable 98.2 98.2 96.5 91.2 96.5

a

DRM, drug resistance mutation; ID, identifier.

airfreight items under the same environmental conditions as passenger luggage. The maximum temperature reached was 40°C, which occurred during a layover in Dubai between flights, while the minimum temperature was 1°C. Thus, the temperatures were rarely above body temperature, a situation that would be expected to cause significant degradation of viral RNA (29, 37, 38). The specimens also did not reach freezing point, thus avoiding freeze-thaw cycles that could also lead to degradation of viral nucleic acids. Shipments that are exposed to more extreme temperatures may result in inferior genotyping efficiency, especially if specimens are collected from patients with lower VL. Most DBS shipments took 4 days to reach the genotyping laboratory in the United States. This relatively short time frame may be due in part to the proximity of the MRC/UVRI laboratory to the international airport in Entebbe, and to daily flights from Entebbe to Dubai, the regional hub for the courier used throughout this study. In other settings, logistical factors such as distance from the international airport, routing options, and flight schedules used by the courier should be considered when planning studies using DBS for HIV genotyping so as to keep the transport time to a minimum. Our study had several limitations. In subgroup analyses, the sample sizes may have been too small to distinguish statistically significant differences in genotyping efficiency between groups. For example, the apparent genotyping efficiency reduction in group 2 for specimens with low VL (89.5% versus 94.7%, n ⫽ 19) could become statistically significant if a larger sample size was used. The environmental temperature conditions during storage and shipments were not as extreme as they might be in other locations. Finally, DBS were prepared from venous blood using pipettes in a controlled laboratory environment. Particular environmental conditions and DBS preparation by less-experienced personnel in other situations could negatively impact genotyping assay performance using DBS. During the second phase of enrollment for this study, we screened individuals who were suspected to be failing ART in order to identify patients who had plasma VL of over 1,000 copies/ ml. Because VL monitoring is limited in Uganda, suspicion of failure (and thus eligibility for switching to a second-line regimen) was based on falling CD4 counts and/or clinical symptoms. All 227 patients screened by this method would have been considered for switching therapy to a second-line regimen, and yet less than

August 2014 Volume 52 Number 8

half (104) had plasma VL of over 1,000 copies/ml. Even with good clinical practice in Uganda, where intensive adherence counseling is the first course of action when therapy failure is suspected, our results suggest the more than half of the possible switches to a second-line regimen would have been unnecessary. A similar finding has also recently been reported from a study of ART treatment-switching practices in six African countries (39). The greatly increased cost of second-line ART that includes a boosted PI suggests that routine VL testing could save costs by avoiding unnecessary treatment switches. WHO has recently updated its guidelines to recommend VL testing where possible, although the financial and logistical feasibility of universal VL testing in resource-limited settings is a significant challenge (40, 41). In conclusion, DBS specimens are a suitable specimen type for HIVDR surveillance and monitoring in resource-limited settings if carefully prepared and stored at ambient temperatures for 2 weeks or less. Frozen DBS specimens can be thawed and shipped at ambient temperature using a standard courier service to genotyping laboratories. DBS should not be stored for longer than 2 weeks at ambient temperature, and a reduction in genotyping efficiency can be expected when finger-prick blood is used for DBS preparation from patients with VL of less than about 50,000 copies/ml. These results help to define the optimal collection, storage, and shipping conditions for DBS specimens for use in HIVDR genotyping and open a new avenue for cost-saving ambient-temperature DBS specimen shipments for HIVDR surveillance and monitoring in resourcelimited settings regardless of the storage conditions before shipment. In the absence of routine individual-level VL monitoring and HIVDR genotyping, population-level surveillance of HIVDR for the purpose of informing national public health policy, including national ART guidelines, is critical (7). Use of DBS specimens for HIVDR surveillance will greatly facilitate nationally representative sampling by enabling inclusion of specimens from patients residing in rural areas. ACKNOWLEDGMENTS We express our sincere thanks to each of the study participants and the staff at the four clinical sites in Uganda. This project has been supported by the President’s Emergency Plan for AIDS Relief (PEPFAR) through U.S. Centers for Disease Control and Prevention, the Bill and Melinda Gates Foundation, and CFAR

jcm.asm.org 2873

Downloaded from http://jcm.asm.org/ on March 8, 2015 by DAHLGREN MEDICAL LIBRARY

Specimen ID

Parry et al.

P30AI42853 and in part by the UK Medical Research Council (MRC) and the UK Department for International Development (DFID) under the MRC/DFID Concordat agreement. C.Y. and J.D. are the inventors in U.S. patent application PCT/ US2012/045523. The findings and conclusions in this report are ours and do not necessarily represent the official position of the Centers for Disease Control and Prevention/the Agency for Toxic Substances and Disease Registry.

17.

18.

REFERENCES

2874

jcm.asm.org

19.

20.

21.

22.

23.

24.

25.

26.

27. 28.

29.

30.

31.

Journal of Clinical Microbiology

Downloaded from http://jcm.asm.org/ on March 8, 2015 by DAHLGREN MEDICAL LIBRARY

1. World Health Organization. 2013. Global update on HIV treatment 2013: results, impact and opportunities. World Health Organization. http: //www.who.int/hiv/pub/progressreports/update2013/en/index.html. Accessed 24 September 2013. 2. Frentz D, Boucher CA, van de Vijver DA. 2012. Temporal changes in the epidemiology of transmission of drug-resistant HIV-1 across the world. AIDS Rev. 14:17–27. http://www.aidsreviews.com/files/2012_14_1_017-0 27.pdf. 3. Stadeli KM, Richman DD. 2013. Rates of emergence of HIV drug resistance in resource-limited settings: a systematic review. Antivir. Ther. 18: 115–123. http://dx.doi.org/10.3851/IMP2437. 4. Sohn AH, Srikantiah P, Sungkanuparph S, Zhang F. 2013. Transmitted HIV drug resistance in Asia. Curr. Opin. HIV AIDS 8:27–33. http://dx.doi .org/10.1097/COH.0b013e32835b804f. 5. Hamers RL, Sigaloff KC, Kityo C, Mugyenyi P, de Wit TF. 2013. Emerging HIV-1 drug resistance after roll-out of antiretroviral therapy in sub-Saharan Africa. Curr. Opin. HIV AIDS 8:19 –26. http://dx.doi.org/10 .1097/COH.0b013e32835b7f94. 6. Panel on Antiretroviral Guidelines for Adults and Adolescents. 2013. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Department of Health and Human Services. http: //aidsinfo.nih.gov/guidelines. Accessed 5 December 2013. 7. Bennett DE, Bertagnolio S, Sutherland D, Gilks CF. 2008. The World Health Organization’s global strategy for prevention and assessment of HIV drug resistance. Antivir. Ther. 13(Suppl 2):1–13. http://www.intmedpress .com/serveFile.cfm?sUID⫽d7e0c91a-2622-4eff-acc3-b45f6e856a67. 8. World Health Organization. 2012. World Health Organization global strategy for the surveillance and monitoring of HIV drug resistance. World Health Organization. http://www.who.int/hiv/pub/drugresistance /drug_resistance_strategy/en/index.html. Accessed 14 February 2012. 9. Plantier JC, Dachraoui R, Lemee V, Gueudin M, Borsa-Lebas F, Caron F, Simon F. 2005. HIV-1 resistance genotyping on dried serum spots. AIDS 19:391–397. http://dx.doi.org/10.1097/01.aids.0000161768.98534.e7. 10. World Health Organization. 2013. Guidance on regulations for the Transport of Infectious Substances 2013-2014. World Health Organization. http://www.who.int/ihr/publications/who_hse_ihr_2012.12/en/. Accessed 9 May 2014. 11. Hamers RL, Smit PW, Stevens W, Schuurman R, Rinke de Wit TF. 2009. Dried fluid spots for HIV type-1 viral load and resistance genotyping: a systematic review. Antivir. Ther. 14:619 – 629. http://www.intmedpress.com /serveFile.cfm?sUID⫽8d7390e5-08b3-45ad-b56d-b7184f85d203. 12. Bertagnolio S, Parkin NT, Jordan M, Brooks J, Garcia-Lerma JG. 2010. Dried blood spots for HIV-1 drug resistance and viral load testing: a review of current knowledge and WHO efforts for global HIV drug resistance surveillance. AIDS Rev. 12:195–208. http://www.aidsreviews.com /files/2010_12_4_195-208.pdf. 13. Buckton AJ. 2008. New methods for the surveillance of HIV drug resistance in the resource poor world. Curr. Opin. Infect. Dis. 21:653– 658. http://dx.doi.org/10.1097/QCO.0b013e3283186d1a. 14. Buckton AJ, Bissett SL, Myers RE, Beddows S, Edwards S, Cane PA, Pillay D. 2008. Development and optimization of an internally controlled dried blood spot assay for surveillance of human immunodeficiency virus type-1 drug resistance. J. Antimicrob. Chemother. 62:1191–1198. http: //dx.doi.org/10.1093/jac/dkn412. 15. Zhou Z, Wagar N, DeVos JR, Rottinghaus E, Diallo K, Nguyen DB, Bassey O, Ugbena R, Wadonda-Kabondo N, McConnell MS, Zulu I, Chilima B, Nkengasong J, Yang C. 2011. Optimization of a low cost and broadly sensitive genotyping assay for HIV-1 drug resistance surveillance and monitoring in resource-limited settings. PLoS One 6:e28184. http: //dx.doi.org/10.1371/journal.pone.0028184. 16. Masciotra S, Garrido C, Youngpairoj AS, McNulty A, Zahonero N, Corral A, Heneine W, de Mendoza C, Garcia-Lerma JG. 2007. High

concordance between HIV-1 drug resistance genotypes generated from plasma and dried blood spots in antiretroviral-experienced patients. AIDS 21:2503–2511. http://dx.doi.org/10.1097/QAD.0b013e3281c618db. Ziemniak C, Mengistu Y, Ruff A, Chen YH, Khaki L, Bedri A, Simen BB, Palumbo P, Eshleman SH, Persaud D. 2011. Use of dried-blood-spot samples and in-house assays to identify antiretroviral drug resistance in HIV-infected children in resource-constrained settings. J. Clin. Microbiol. 49:4077– 4082. http://dx.doi.org/10.1128/JCM.01004-11. Johannessen A, Garrido C, Zahonero N, Naman E, de Mendoza C. 2011. HIV-1 drug resistance testing from dried blood spots collected in rural Tanzania using the ViroSeq HIV-1 Genotyping System. J. Antimicrob. Chemother. 66:260 –264. http://dx.doi.org/10.1093/jac/dkq433. Monleau M, Butel C, Delaporte E, Boillot F, Peeters M. 2010. Effect of storage conditions of dried plasma and blood spots on HIV-1 RNA quantification and PCR amplification for drug resistance genotyping. J. Antimicrob. Chemother. 65:1562–1566. http://dx.doi.org/10.1093/jac/dkq205. Hallack R, Doherty LE, Wethers JA, Parker MM. 2008. Evaluation of dried blood spot specimens for HIV-1 drug-resistance testing using the Trugene HIV-1 genotyping assay. J. Clin. Virol. 41:283–287. http://dx.doi .org/10.1016/j.jcv.2007.12.011. Youngpairoj AS, Masciotra S, Garrido C, Zahonero N, de Mendoza C, Garcia-Lerma JG. 2008. HIV-1 drug resistance genotyping from dried blood spots stored for 1 year at 4°C. J. Antimicrob. Chemother. 61:1217– 1220. http://dx.doi.org/10.1093/jac/dkn100. Parkin N, de Mendoza C, Schuurman R, Jennings C, Bremer J, Jordan MR, Bertagnolio S. 2012. Evaluation of in-house genotyping assay performance using dried blood spot specimens in the Global World Health Organization laboratory network. Clin. Infect. Dis. 54(Suppl 4):S273– S279. http://dx.doi.org/10.1093/cid/cir982. Steegen K, Luchters S, Demecheleer E, Dauwe K, Mandaliya K, Jaoko W, Plum J, Temmerman M, Verhofstede C. 2007. Feasibility of detecting human immunodeficiency virus type 1 drug resistance in DNA extracted from whole blood or dried blood spots. J. Clin. Microbiol. 45:3342–3351. http://dx.doi.org/10.1128/JCM.00814-07. McNulty A, Jennings C, Bennett D, Fitzgibbon J, Bremer JW, Ussery M, Kalish ML, Heneine W, Garcia-Lerma JG. 2007. Evaluation of dried blood spots for human immunodeficiency virus type 1 drug resistance testing. J. Clin. Microbiol. 45:517–521. http://dx.doi.org/10.1128/JCM .02016-06. Rottinghaus EK, Ugbena R, Diallo K, Bassey O, Azeez A, Devos J, Zhang G, Aberle-Grasse J, Nkengasong J, Yang C. 2012. Dried blood spot specimens are a suitable alternative sample type for HIV-1 viral load measurement and drug resistance genotyping in patients receiving firstline antiretroviral therapy. Clin. Infect. Dis. 54:1187–1195. http://dx.doi .org/10.1093/cid/cis015. Bertagnolio S, Soto-Ramirez L, Pilon R, Rodriguez R, Viveros M, Fuentes L, Harrigan PR, Mo T, Sutherland D, Sandstrom P. 2007. HIV-1 drug resistance surveillance using dried whole blood spots. Antivir. Ther. 12:107–113. http://www.intmedpress.com/serveFile.cfm?sUID⫽c2 e29fe6-5ce2-422a-b3e0-40432e135435. Hearps AC, Ryan CE, Morris LM, Plate MM, Greengrass V, Crowe SM. 2010. Stability of dried blood spots for HIV-1 drug resistance analysis. Curr. HIV Res. 8:134 –140. http://dx.doi.org/10.2174/157016210790442740. Ziemniak C, George-Agwu A, Moss WJ, Ray SC, Persaud D. 2006. A sensitive genotyping assay for detection of drug resistance mutations in reverse transcriptase of HIV-1 subtypes B and C in samples stored as dried blood spots or frozen RNA extracts. J. Virol. Methods 136:238 –247. http: //dx.doi.org/10.1016/j.jviromet.2006.05.030. García-Lerma JG, McNulty A, Jennings C, Huang D, Heneine W, Bremer JW. 2009. Rapid decline in the efficiency of HIV drug resistance genotyping from dried blood spots (DBS) and dried plasma spots (DPS) stored at 37 degrees C and high humidity. J. Antimicrob. Chemother. 64:33–36. http://dx.doi.org/10.1093/jac/dkp150. Monleau M, Montavon C, Laurent C, Segondy M, Montes B, Delaporte E, Boillot F, Peeters M. 2009. Evaluation of different RNA extraction methods and storage conditions of dried plasma or blood spots for human immunodeficiency virus type 1 RNA quantification and PCR amplification for drug resistance testing. J. Clin. Microbiol. 47:1107–1118. http://dx .doi.org/10.1128/JCM.02255-08. Yang C, McNulty A, Diallo K, Zhang J, Titanji B, Kassim S, WadondaKabondo N, Aberle-Grasse J, Kibuka T, Ndumbe PM, Vedapuri S, Zhou Z, Chilima B, Nkengasong JN. 2010. Development and application of a broadly sensitive dried-blood-spot-based genotyping assay for global

Field Study of DBS on HIV-1 Drug Resistance Genotyping

32.

33. 34.

36.

37.

August 2014 Volume 52 Number 8

38.

39.

40.

41.

blood during specimen handling and storage prior to amplification by NASBA-QT. J. Virol. 67:199 –207. http://www.sciencedirect.com/science /article/pii/S0166093497000979. Amellal B, Murphy R, Maiga A, Brucker G, Katlama C, Calvez V, Marcelin AG. 2008. Stability of HIV RNA in plasma specimens stored at different temperatures. HIV Med. 9:790 –793. http://dx.doi.org/10.1111/j .1468-1293.2008.00632.x. Sigaloff KC, Hamers RL, Wallis CL, Kityo C, Siwale M, Ive P, Botes ME, Mandaliya K, Wellington M, Osibogun A, Stevens WS, van Vugt M, de Wit TF; PharmAccess African Studies to Evaluate Resistance (PASER). 2011. Unnecessary antiretroviral treatment switches and accumulation of HIV resistance mutations; two arguments for viral load monitoring in Africa. J. Acquir. Immune Defic. Syndr. 58:23–31. http://dx.doi.org/10 .1097/QAI.0b013e318227fc34. Hirnschall G, Harries AD, Easterbrook PJ, Doherty MC, Ball A. 2013. The next generation of the World Health Organization’s global antiretroviral guidance. J. Int. AIDS Soc. 16:18757. http://dx.doi.org/10.7448/IAS .16.1.18757. World Health Organization. 2013. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: Recommendations for a public health approach. World Health Organization. http://apps.who.int/iris/bitstream/10665/85321/1/9789241505727_eng .pdf. Accessed 23 September 2013.

jcm.asm.org 2875

Downloaded from http://jcm.asm.org/ on March 8, 2015 by DAHLGREN MEDICAL LIBRARY

35.

surveillance of HIV-1 drug resistance. J. Clin. Microbiol. 48:3158 –3164. http://dx.doi.org/10.1128/JCM.00564-10. Woods CK, Brumme CJ, Liu TF, Chui CK, Chu AL, Wynhoven B, Hall TA, Trevino C, Shafer RW, Harrigan PR. 2012. Automating HIV drug resistance genotyping with RECall, a freely accessible sequence analysis tool. J. Clin. Microbiol. 50:1936 –1942. http://dx.doi.org/10.1128/JCM.06689-11. Liu TF, Shafer RW. 2006. Web resources for HIV type 1 genotypicresistance test interpretation. Clin. Infect. Dis. 42:1608 –1618. http://dx .doi.org/10.1086/503914. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28:2731–2739. http://dx.doi.org/10.1093/molbev/msr121. Benjamini Y, Hochberg Y. 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B 57:289 –300. http://www.jstor.org/stable/pdfplus/2346101.pdf. Fajardo E, Metcalf CA, Chaillet P, Aleixo L, Pannus P, Panunzi I, Trivino L, Ellman T, Likaka A, Mwenda R. 2014. Prospective evaluation of diagnostic accuracy of dried blood spots from finger prick samples for determination of HIV-1 load with the NucliSENS Easy-Q HIV-1 version 2.0 assay in Malawi. J. Clin. Microbiol. 52:1343–1351. http://dx.doi.org/10 .1128/JCM.03519-13. Bruisten SM, Oudshoorn P, van Swieten P, Boeser-Nunnink B, van Aarle P, Tondreau SP, Cuypers HTM. 1997. Stability of HIV-1 RNA in