TY - JOUR
AU - Manuel, Haschke,
AB - Abstract Objectives We evaluated whether dried blood spots (DBS) are suitable to monitor combined ART when samples are collected in rural Tanzania and transported over a long distance to a specialized bioanalytical laboratory. Methods Plasma and DBS samples were collected in Tanzania from study patients treated with nevirapine, efavirenz or lopinavir. In addition, plasma, whole blood and DBS samples were obtained from a cohort of HIV patients at the site of the bioanalytical laboratory in Switzerland. DBS samples were analysed using a fully automated LC-MS/MS method. Results Comparison of DBS versus plasma concentrations of samples obtained from the bridging study in Switzerland indicated an acceptable bias only for nevirapine (18.4%), whereas for efavirenz and lopinavir a pronounced difference of −47.4% and −48.1% was found, respectively. Adjusting the DBS concentrations by the haematocrit and the fraction of drug bound to plasma proteins removed this bias [efavirenz +9.4% (−6.9% to +25.7%), lopinavir +2.2% (−20.0% to +24.2%)]. Storage and transportation of samples from Tanzania to Switzerland did not affect the good agreement between plasma and DBS for nevirapine [–2.9% (−34.7% to +29.0%)] and efavirenz [–9.6% (−42.9% to +23.8%)]. For lopinavir, however, adjusted DBS concentrations remained considerably below [–32.8% (−70.4% to +4.8%)] corresponding plasma concentrations due to decay of lopinavir in DBS obtained under field conditions. Conclusions Our field study shows that the DBS technique is a suitable tool for therapeutic drug monitoring in resource-poor regions; however, sample stability remains an issue for certain analytes and therefore needs special consideration. Introduction Therapeutic drug monitoring (TDM) is an inherent part of medical care in industrialized countries.1 It is an important tool for physicians not only to guide dose adaptation but also to assess patients’ adherence. However, well-equipped and specialized laboratories are required, which often are not available in resource-poor regions.2–4 To promote further the TDM in developing countries, blood sampling techniques are needed that allow collection of patient samples easily and at low-cost under field conditions.5 Moreover, the samples should be stable under local climate conditions, as uninterrupted cold-chains are often not available.6 Using dried blood spots (DBS) offers many advantages in this context. Collecting a few capillary blood drops from the fingertip is minimally invasive and does not require a trained phlebotomist.7 Moreover, the drying of blood drops on filter paper in many cases improves sample stability compared with the use of wet matrices, which often allows storage at room temperature and reduces the biohazard risk.8 The DBS technique was introduced in the 1960s and is nowadays still routinely used in neonatal screening, but less frequently in highly regulated areas, e.g. in drug development or for routine TDM of small molecules.9 Literature data are usually based on plasma and not on capillary blood concentrations, so that the relationship between the two matrices has to be understood for the correct interpretation of DBS measurements. In addition, bioanalytical aspects, e.g. lower assay sensitivity, analytical bias caused by variable haematocrit values and the complexity of extraction, limit the widespread acceptance of the DBS technique.10 Nevertheless, recent technical advances in the automation of DBS extraction and analysis have led to more time-efficient, robust and sensitive measurements.10,11 TDM is a valuable tool to improve efficacy and safety in HIV therapy by identifying patients with toxic, subtherapeutic or appropriate drug concentrations.12 To maintain sufficient viral suppression, which is linked to increased survival and reduced morbidity, patients must adhere to ≥95% of the prescribed dosages.13–16 In addition, insufficient drug exposure increases the risk of HIV progression and the development of drug resistance. Worldwide an estimated 37 million people are infected with HIV; the highest prevalence can be found in Sub-Saharan Africa.17 Considering the previously mentioned advantages of DBS, the technique could help to establish TDM as an important tool in the management of patients with HIV living in rural areas with high HIV prevalence. The aim of this study was to evaluate whether DBS are suitable to monitor combined ART, when samples are collected in rural Tanzania and transported over a long distance to a specialized bioanalytical laboratory. DBS, whole blood and plasma TDM samples were acquired at the site of the bioanalytical laboratory (the bridging study) from patients treated with either the NNRTIs efavirenz or nevirapine, or the PI lopinavir. In addition, DBS and plasma samples were collected in Tanzania (the field study18) and transported to Switzerland. DBS samples were extracted using a fully automated system to facilitate the analysis of the large number of study samples. Patients and methods Ethics Ethics approval for the field study was received from the Institutional Review Board of the Ifakara Health Institute (no. IHI 28-2013), the Tanzanian National Institute of Medical Research, Dar es Salaam, Tanzania (no. NIMR/HQ/R.8a/V01. IX/I762) and the Tanzanian Commission for Science & Technology (no. 2014-276-NA-2014-195). For the bridging study, anonymized samples from patients enrolled in the Swiss HIV Cohort Study were used. Ethics approval was received from the ethics committee of Northwest and Central Switzerland (no. EK 76/13). The bridging study was conducted from April until June 2013, whereas the field study samples were collected from October 2013 to September 2014. Informed consent was obtained from all participants.18 Study design Bridging study Sixty HIV-positive patients followed at the University Hospital Basel, Switzerland, who received either nevirapine (n = 20), efavirenz (n = 20) or lopinavir (n = 20) were enrolled in the bridging study. Patients were treated with a median oral dose of 400 mg nevirapine (range 200–400 mg), 600 mg efavirenz (range 200–800 mg) or 800 mg lopinavir (range 400–800 mg). Plasma, whole blood and DBS samples were collected simultaneously during a routine clinical visit. Whole blood was sampled by venepuncture from the antecubital vein into EDTA-containing tubes. Plasma was obtained by centrifugation using standard settings. DBS samples were collected via capillary puncture from the tip of the middle or ring finger. The first blood drop was discarded and subsequent drops of blood were transferred onto DBS filter paper cards (226 grade; PerkinElmer, Waltham, MA, USA). DBS samples were dried for 2 h at room temperature and placed in small plastic bags containing a desiccant. All samples were stored at −20°C until analysis to ensure sample stability. Field study DBS and the matching plasma samples (nevirapine n = 182, efavirenz n = 481 or lopinavir n = 64) were collected within the framework of an adherence assessment study in Tanzania, including 299 patients.18 Between one and three samples were collected per patient on different occasions. Patients were treated with a single daily dose of 600 mg efavirenz, two daily doses of 200 mg nevirapine, or lopinavir/ritonavir 400/100 mg twice daily. Plasma and DBS samples were collected as described above. Plasma samples were stored at −20°C and shipped on dry ice to the bioanalytical laboratory (Basel, Switzerland). DBS samples were stored in the fridge and shipped at room temperature. At the bioanalytical laboratory, plasma and DBS samples were stored at −80°C and analysed after all samples had been received to minimize inter-batch variability. Bioanalytical analyses The antiretrovirals (ARVs) were analysed in DBS using a recently developed fully automated LC-MS/MS method.11 Plasma and DBS samples were analysed using methods validated according to the FDA guideline on bioanalytical method validation.19 A qualified method was used for the analysis of whole blood samples (see Table S1, available as Supplementary data at JAC Online). More details about the method and the performance of the analyses are given in the Supplementary data. A precision of 15% and a mean accuracy of 85%–115% were accepted in this study. Table 1. Nevirapine, efavirenz and lopinavir concentrations [median (IQR)] determined in plasma, DBS and whole blood Nevirapine Efavirenz Lopinavir Matrix bridging study (Switzerland) field study (Tanzania) bridging study (Switzerland) field study (Tanzania) bridging study (Switzerland) field study (Tanzania) Plasma 5305 (3495–6708) 7240 (5050–10 125) 2705 (1963–3468) 2630 (1895–4225) 9245 (6720–14 125) 8870 (6645–11 350) DBS 6800 (4255–7598) 7325 (5070–10 625) 1575 (1295–2250) 1530 (1105–2505) 6120 (4563–8063) 4005 (3103–5395) Whole blood 5650 (3713–7455) ND 1580 (1253–2230) ND 5840 (4108–7318) ND Nevirapine Efavirenz Lopinavir Matrix bridging study (Switzerland) field study (Tanzania) bridging study (Switzerland) field study (Tanzania) bridging study (Switzerland) field study (Tanzania) Plasma 5305 (3495–6708) 7240 (5050–10 125) 2705 (1963–3468) 2630 (1895–4225) 9245 (6720–14 125) 8870 (6645–11 350) DBS 6800 (4255–7598) 7325 (5070–10 625) 1575 (1295–2250) 1530 (1105–2505) 6120 (4563–8063) 4005 (3103–5395) Whole blood 5650 (3713–7455) ND 1580 (1253–2230) ND 5840 (4108–7318) ND ND, not determined. All concentrations are in ng/mL. Table 1. Nevirapine, efavirenz and lopinavir concentrations [median (IQR)] determined in plasma, DBS and whole blood Nevirapine Efavirenz Lopinavir Matrix bridging study (Switzerland) field study (Tanzania) bridging study (Switzerland) field study (Tanzania) bridging study (Switzerland) field study (Tanzania) Plasma 5305 (3495–6708) 7240 (5050–10 125) 2705 (1963–3468) 2630 (1895–4225) 9245 (6720–14 125) 8870 (6645–11 350) DBS 6800 (4255–7598) 7325 (5070–10 625) 1575 (1295–2250) 1530 (1105–2505) 6120 (4563–8063) 4005 (3103–5395) Whole blood 5650 (3713–7455) ND 1580 (1253–2230) ND 5840 (4108–7318) ND Nevirapine Efavirenz Lopinavir Matrix bridging study (Switzerland) field study (Tanzania) bridging study (Switzerland) field study (Tanzania) bridging study (Switzerland) field study (Tanzania) Plasma 5305 (3495–6708) 7240 (5050–10 125) 2705 (1963–3468) 2630 (1895–4225) 9245 (6720–14 125) 8870 (6645–11 350) DBS 6800 (4255–7598) 7325 (5070–10 625) 1575 (1295–2250) 1530 (1105–2505) 6120 (4563–8063) 4005 (3103–5395) Whole blood 5650 (3713–7455) ND 1580 (1253–2230) ND 5840 (4108–7318) ND ND, not determined. All concentrations are in ng/mL. Stability of the ARV was assessed in DBS under accelerated storage conditions mimicking the climate in Tanzania. Stability was evaluated after 1 day and after 1, 2, 3 and 4 weeks of storage at 40°C and 75% relative humidity (rH). Five replicates (500 ng/mL) were analysed and compared with a set of freshly prepared samples. Samples were considered stable if the change in concentration was <15%. Statistical analyses The level of agreement of the ARV concentrations measured in DBS and plasma was assessed using Bland–Altman analysis. The measured concentrations in the different biofluids were compared pairwise by plotting the percentage differences ( %difference=concentration in fluid 1-concentration in fluid 2mean concentration×100) against their mean values. The mean deviation and the 95% limits of agreement were calculated. DBS concentrations were adjusted with the patient’s haematocrit and with the fraction of drug bound to plasma protein (fbpp) as proposed by Kromdijk et al.20 ( [Analyte]Plasma=[Analyte]DBS(1-HCT)×fbpp). The haematocrit was determined for each patient while the fbpp used for nevirapine, efavirenz and lopinavir was 0.6, 0.995 and 0.98, respectively.20,21 Another algorithm for plasma and DBS concentrations was evaluated ( [Analyte]Plasma=[Analyte]DBS1-HCT+KRBC/plasma × HCT), which considers the haematocrit as well as the partitioning of the drug between plasma and red blood cells (KRBC/plasma).22,KRBC/plasma was calculated based on the measured haematocrit, and the whole blood and plasma concentrations from the participants of the bridging study. The mean KRBC/plasma (n = 20) was then determined for each analyte by rearranging the equation published by Jager et al.22 A cross-validation of the concentrations obtained in the different study sample matrices was performed according to bioanalytical method validation guidelines, whereby at least 67% of the samples were required to be within the ±20% limits.23,24 Statistical analyses were carried out with GraphPad Prism 7 (La Jolla, CA, USA). A subtherapeutic drug concentration was defined as any concentration below the 2.5th percentile of published population-based pharmacokinetic models for the single daily dose of 600 mg efavirenz,25 two daily doses of 200 mg nevirapine26 and lopinavir/ritonavir 400/100 mg twice daily.27 Results Participant characteristics Sixty participants were enrolled in the bridging study, with an equal number of patients receiving treatment either with nevirapine, efavirenz or lopinavir. Of the participants, 65% were male and the median age was 50 years (range 27–73 years). The median haematocrit was 42% and ranged between 30% and 52%. Samples were withdrawn at a median of 11 h (range 1–36 h) post treatment. The characteristics of the patients enrolled in the field study were summarized by Erb et al.18 In brief, samples were collected within 2–24 h post-dosing, which was comparable to the bridging study. Haematocrits were generally lower than in the bridging study with a median of 35.9% and a range from 10.9% to 50.8%. Concentrations determined in plasma, whole blood and DBS are summarized in Table 1. RBC partitioning KRBC/plasma of the ARV was calculated based on plasma and blood data obtained within the bridging study. Concentrations measured in blood and plasma following nevirapine treatment resulted in a mean KRBC/plasma of 1.15 (95% CI 1.07–1.24). In comparison, the KRBC/plasma for efavirenz and lopinavir indicated that only a small amount of efavirenz [0.13 (95% CI 0.08–0.18)] and almost no lopinavir [–0.039 (95% CI −0.10–0.02)] distributed into the red blood cell (RBC). Agreement between plasma and blood samples The percentage differences between blood and plasma concentrations for nevirapine, efavirenz and lopinavir were consistent without a trend across the monitored concentration range. Blood versus plasma concentrations indicated only a small bias of +5.7% for nevirapine (Table 2). In contrast, comparison of efavirenz and lopinavir blood and plasma concentrations displayed a pronounced difference of −47.8% and −55.2%, respectively. The 95% limits of agreement for all three drugs were narrow with a range of less than ±25%. For nevirapine, no correction between plasma and blood was necessary, because the difference of all samples is already within the ±20% limits (Table 2). Efavirenz and lopinavir blood concentrations had to be adjusted by the haematocrit and the KRBC/plasma or fbpp to estimate correctly the concentrations in plasma, due to the low amount of drug distributing into the RBCs. Both correction modalities significantly improved the agreement between plasma and blood concentrations so that the mean percentage difference was <9% and at least 95% of the samples were within the ±20% limits (Table 2). Table 2. Percentage differences (95% limits of agreement) determined for nevirapine, efavirenz and lopinavir between different matrices Bridging study (Switzerland) Field study (Tanzania) Analyte/correction plasma and blood no. within ±20% limits (%) plasma and DBS no. within ±20% limits (%) plasma and DBS no. within ±20% limits (%) Nevirapine  no correction +5.7 (−6.1 to +17.7) 20 (100) +18.4 (+2.1 to +34.8) 12 (60) +3.6 (−25.0 to +32.1) 158 (87)  corrected by HCT and fbpp +7.7 (−12.3 to +27.8) 18 (90) +20.3 (−6.3 to +46.9) 9 (45) −2.9 (−34.7 to +29.0) 142 (78)  corrected by HCT and KRBC/plasmaa −0.4 (−12.6 to +11.8) 20 (100) +12.4 (−3.7 to +28.5) 18 (90) −1.7 (−30.3 to +26.8) 160 (88) Efavirenz  no correction −47.8 (−65.2 to −30.4) 0 (0) −47.4 (−63.9 to −30.9) 0 (0) −52.0 (−84.8 to −19.3) 2 (0.4)  corrected by HCT and fbpp +9.0 (−6.6 to +24.4) 19 (95) +9.4 (−6.9 to +25.7) 20 (100) −9.6 (−42.9 to +23.8) 364 (76)  corrected by HCT and KRBC/plasmaa −0.3 (−16.0 to +15.3) 20 (100) +0.1 (−15.7 to +16.0) 20 (100) −16.1 (−48.9 to +16.8) 379 (58) Lopinavir  no correction −55.2 (−77.6 to −33.0) 0 (0) −48.1 (−69.0 to −27.2) 0 (0) −72.3 (−107.5 to −37.2) 0 (0)  corrected by HCT and fbpp −5.6 (−23.1 to +11.9) 19 (95) +2.2 (−20.0 to 24.2) 19 (95) −32.8 (−70.4 to +4.8) 16 (25)  corrected by HCT and KRBC/plasmaa −3.6 (−21.1 to 13.9) 20 (100) +4.2 (−17.8 to + 26.2) 18 (90) −31.0 (−68.6 to +6.7) 17 (27) Bridging study (Switzerland) Field study (Tanzania) Analyte/correction plasma and blood no. within ±20% limits (%) plasma and DBS no. within ±20% limits (%) plasma and DBS no. within ±20% limits (%) Nevirapine  no correction +5.7 (−6.1 to +17.7) 20 (100) +18.4 (+2.1 to +34.8) 12 (60) +3.6 (−25.0 to +32.1) 158 (87)  corrected by HCT and fbpp +7.7 (−12.3 to +27.8) 18 (90) +20.3 (−6.3 to +46.9) 9 (45) −2.9 (−34.7 to +29.0) 142 (78)  corrected by HCT and KRBC/plasmaa −0.4 (−12.6 to +11.8) 20 (100) +12.4 (−3.7 to +28.5) 18 (90) −1.7 (−30.3 to +26.8) 160 (88) Efavirenz  no correction −47.8 (−65.2 to −30.4) 0 (0) −47.4 (−63.9 to −30.9) 0 (0) −52.0 (−84.8 to −19.3) 2 (0.4)  corrected by HCT and fbpp +9.0 (−6.6 to +24.4) 19 (95) +9.4 (−6.9 to +25.7) 20 (100) −9.6 (−42.9 to +23.8) 364 (76)  corrected by HCT and KRBC/plasmaa −0.3 (−16.0 to +15.3) 20 (100) +0.1 (−15.7 to +16.0) 20 (100) −16.1 (−48.9 to +16.8) 379 (58) Lopinavir  no correction −55.2 (−77.6 to −33.0) 0 (0) −48.1 (−69.0 to −27.2) 0 (0) −72.3 (−107.5 to −37.2) 0 (0)  corrected by HCT and fbpp −5.6 (−23.1 to +11.9) 19 (95) +2.2 (−20.0 to 24.2) 19 (95) −32.8 (−70.4 to +4.8) 16 (25)  corrected by HCT and KRBC/plasmaa −3.6 (−21.1 to 13.9) 20 (100) +4.2 (−17.8 to + 26.2) 18 (90) −31.0 (−68.6 to +6.7) 17 (27) fbpp, fraction of drug bound to plasma protein; KRBC/plasma, partition of the drug between plasma and RBCs; HCT, haematocrit. a KRBC/plasma of 1.15, 0.13 and 0 was used for nevirapine, efavirenz and lopinavir, respectively. Values were determined from in vivo blood and plasma drug distribution data. Table 2. Percentage differences (95% limits of agreement) determined for nevirapine, efavirenz and lopinavir between different matrices Bridging study (Switzerland) Field study (Tanzania) Analyte/correction plasma and blood no. within ±20% limits (%) plasma and DBS no. within ±20% limits (%) plasma and DBS no. within ±20% limits (%) Nevirapine  no correction +5.7 (−6.1 to +17.7) 20 (100) +18.4 (+2.1 to +34.8) 12 (60) +3.6 (−25.0 to +32.1) 158 (87)  corrected by HCT and fbpp +7.7 (−12.3 to +27.8) 18 (90) +20.3 (−6.3 to +46.9) 9 (45) −2.9 (−34.7 to +29.0) 142 (78)  corrected by HCT and KRBC/plasmaa −0.4 (−12.6 to +11.8) 20 (100) +12.4 (−3.7 to +28.5) 18 (90) −1.7 (−30.3 to +26.8) 160 (88) Efavirenz  no correction −47.8 (−65.2 to −30.4) 0 (0) −47.4 (−63.9 to −30.9) 0 (0) −52.0 (−84.8 to −19.3) 2 (0.4)  corrected by HCT and fbpp +9.0 (−6.6 to +24.4) 19 (95) +9.4 (−6.9 to +25.7) 20 (100) −9.6 (−42.9 to +23.8) 364 (76)  corrected by HCT and KRBC/plasmaa −0.3 (−16.0 to +15.3) 20 (100) +0.1 (−15.7 to +16.0) 20 (100) −16.1 (−48.9 to +16.8) 379 (58) Lopinavir  no correction −55.2 (−77.6 to −33.0) 0 (0) −48.1 (−69.0 to −27.2) 0 (0) −72.3 (−107.5 to −37.2) 0 (0)  corrected by HCT and fbpp −5.6 (−23.1 to +11.9) 19 (95) +2.2 (−20.0 to 24.2) 19 (95) −32.8 (−70.4 to +4.8) 16 (25)  corrected by HCT and KRBC/plasmaa −3.6 (−21.1 to 13.9) 20 (100) +4.2 (−17.8 to + 26.2) 18 (90) −31.0 (−68.6 to +6.7) 17 (27) Bridging study (Switzerland) Field study (Tanzania) Analyte/correction plasma and blood no. within ±20% limits (%) plasma and DBS no. within ±20% limits (%) plasma and DBS no. within ±20% limits (%) Nevirapine  no correction +5.7 (−6.1 to +17.7) 20 (100) +18.4 (+2.1 to +34.8) 12 (60) +3.6 (−25.0 to +32.1) 158 (87)  corrected by HCT and fbpp +7.7 (−12.3 to +27.8) 18 (90) +20.3 (−6.3 to +46.9) 9 (45) −2.9 (−34.7 to +29.0) 142 (78)  corrected by HCT and KRBC/plasmaa −0.4 (−12.6 to +11.8) 20 (100) +12.4 (−3.7 to +28.5) 18 (90) −1.7 (−30.3 to +26.8) 160 (88) Efavirenz  no correction −47.8 (−65.2 to −30.4) 0 (0) −47.4 (−63.9 to −30.9) 0 (0) −52.0 (−84.8 to −19.3) 2 (0.4)  corrected by HCT and fbpp +9.0 (−6.6 to +24.4) 19 (95) +9.4 (−6.9 to +25.7) 20 (100) −9.6 (−42.9 to +23.8) 364 (76)  corrected by HCT and KRBC/plasmaa −0.3 (−16.0 to +15.3) 20 (100) +0.1 (−15.7 to +16.0) 20 (100) −16.1 (−48.9 to +16.8) 379 (58) Lopinavir  no correction −55.2 (−77.6 to −33.0) 0 (0) −48.1 (−69.0 to −27.2) 0 (0) −72.3 (−107.5 to −37.2) 0 (0)  corrected by HCT and fbpp −5.6 (−23.1 to +11.9) 19 (95) +2.2 (−20.0 to 24.2) 19 (95) −32.8 (−70.4 to +4.8) 16 (25)  corrected by HCT and KRBC/plasmaa −3.6 (−21.1 to 13.9) 20 (100) +4.2 (−17.8 to + 26.2) 18 (90) −31.0 (−68.6 to +6.7) 17 (27) fbpp, fraction of drug bound to plasma protein; KRBC/plasma, partition of the drug between plasma and RBCs; HCT, haematocrit. a KRBC/plasma of 1.15, 0.13 and 0 was used for nevirapine, efavirenz and lopinavir, respectively. Values were determined from in vivo blood and plasma drug distribution data. Agreement between plasma and DBS samples Overall, the percentage differences between plasma and DBS samples did not display a trend across the observed concentrations (Figures 1 and 2). Figure 1. View largeDownload slide Bland–Altman plots of nevirapine, efavirenz and lopinavir concentrations measured in plasma and DBS of samples collected within the bridging study in Switzerland. Values with a difference of less than ±20% between the two matrices are shown using filled symbols; values with larger differences are shown using open symbols. Dotted line (red) shows the mean difference between plasma and DBS samples, the dashed lines the ±20% limits. White plot area illustrates the 95% limits of agreement. DBS concentrations are uncorrected (left column) or adjusted by haematocrit (HCT) value and protein binding (fbpp, right column). This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 1. View largeDownload slide Bland–Altman plots of nevirapine, efavirenz and lopinavir concentrations measured in plasma and DBS of samples collected within the bridging study in Switzerland. Values with a difference of less than ±20% between the two matrices are shown using filled symbols; values with larger differences are shown using open symbols. Dotted line (red) shows the mean difference between plasma and DBS samples, the dashed lines the ±20% limits. White plot area illustrates the 95% limits of agreement. DBS concentrations are uncorrected (left column) or adjusted by haematocrit (HCT) value and protein binding (fbpp, right column). This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 2. View largeDownload slide Bland–Altman plots of nevirapine, efavirenz and lopinavir concentrations measured in plasma and DBS of samples collected within the field study in Tanzania. Samples were collected at one to three occasions for each patient. Values with a difference of less than ±20% between the two matrices are shown using filled symbols; values with larger differences are shown using open symbols. Dotted line (red) shows the mean difference between plasma and DBS samples, the dashed lines the ±20% limits. White plot area illustrates the 95% limits of agreement. DBS concentrations were not corrected (left column) or adjusted by haematocrit (HCT) value and protein binding (fbpp, right column). Y-axes were kept small for better visualization of the data; thus, a few data points were outside the axis limits (left column: nevirapine n = 1 and efavirenz n = 2; right column: nevirapine n = 1 and efavirenz n = 3). This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 2. View largeDownload slide Bland–Altman plots of nevirapine, efavirenz and lopinavir concentrations measured in plasma and DBS of samples collected within the field study in Tanzania. Samples were collected at one to three occasions for each patient. Values with a difference of less than ±20% between the two matrices are shown using filled symbols; values with larger differences are shown using open symbols. Dotted line (red) shows the mean difference between plasma and DBS samples, the dashed lines the ±20% limits. White plot area illustrates the 95% limits of agreement. DBS concentrations were not corrected (left column) or adjusted by haematocrit (HCT) value and protein binding (fbpp, right column). Y-axes were kept small for better visualization of the data; thus, a few data points were outside the axis limits (left column: nevirapine n = 1 and efavirenz n = 2; right column: nevirapine n = 1 and efavirenz n = 3). This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. The mean percentage difference of nevirapine plasma and DBS concentrations was higher in the bridging study (+18.4%) compared with the field study (+3.6%). Thus, only 60% of the nevirapine samples were within the ±20% limits in the bridging study, whereas 87% of the field study samples were within these boundaries (Figures 1 and 2). Correction of the DBS samples by the haematocrit and KRBC/plasma improved the agreement between plasma and DBS samples of the bridging study samples so that 90% of the samples were within ±20%. However, the agreement was not improved when adjusting the DBS concentrations by the haematocrit and the protein binding. Efavirenz concentrations found in DBS were very similar to levels measured in whole blood, but amounted to only about half of the concentration found in plasma. The mean percentage differences of the DBS and plasma samples collected within the bridging (−47.4%) and the field study (−52.0%) were similar, but samples from Tanzania were more scattered, leading to wider 95% limits of agreement. Almost none of the plasma and DBS samples exhibited a difference of <20% if no correction was applied to DBS samples. Correction by the haematocrit and the protein binding or the KRBC/plasma significantly improved the agreement between plasma and DBS measurements. Both correction functions nullified the bias between plasma and DBS samples in case of the bridging study so that no samples had a >20% deviation and the mean percentage difference was <10%. The agreement between efavirenz plasma and DBS samples collected in Tanzania was improved by both correction modalities as well, resulting in a mean percentage difference of less than −16.1%. Seventy-six percent of the samples were within the specified limits of ±20% after adjusting DBS concentrations by the haematocrit and the protein binding, and thus passed the criterion of 67% for cross-validation. Lopinavir concentrations measured in DBS were more than 50% lower than in plasma, similar to the observation accounted for in efavirenz samples. Furthermore, the 95% limits of agreement between plasma and DBS were similar compared with plasma and blood samples. Adjusting the lopinavir concentrations in DBS samples from the bridging study based on their haematocrit and the protein binding or KRBC/plasma resulted in a substantial improvement of the agreement. More than 90% of the samples were within 20% limits and the mean percentage difference was less than +5%. However, lopinavir concentrations in DBS samples from Tanzania were still ∼30% lower than in plasma after correction, and only a quarter of the samples were within ±20%. Overall, 4.2% (n = 20), 2.2% (n = 4) and 6.3% (n = 4) of the measured efavirenz, nevirapine and lopinavir plasma samples, respectively, were in a subtherapeutic range (Figure S1). The fractions of subtherapeutic samples increased to 14.0% (n = 67) for efavirenz and to 10.9% (n = 7) for lopinavir if calculations were based on uncorrected DBS measurements. For nevirapine, DBS analysis revealed the same percentage of subtherapeutic samples as for plasma analysis. Interpretation of the TDM samples using DBS instead of plasma was comparable if the DBS concentrations were corrected by the haematocrit and the protein binding, so that 5.2% (plasma 4.2%) of the efavirenz, 2.7% (plasma 2.2%) of the nevirapine and 6.3% (plasma 6.3%) of the lopinavir concentrations were judged as subtherapeutic. Stability of DBS samples The stability of the ARV drugs was tested under accelerated storage conditions (40°C, 75% rH) because the percentage differences of lopinavir DBS and plasma samples from Tanzania were unexpectedly large in comparison with the data obtained within the bridging study (Figure 3). Figure 3. View largeDownload slide Stability of nevirapine, efavirenz and lopinavir under accelerated storage conditions at 40°C and 75% rH over 4 weeks. Five replicates were analysed after 1 day and after 1, 2, 3 and 4 weeks. Freshly prepared samples were used as comparators to assess the stability of the ARV drugs. Circles indicate the calculated mean percentage stability and the error bars are standard deviations. Figure 3. View largeDownload slide Stability of nevirapine, efavirenz and lopinavir under accelerated storage conditions at 40°C and 75% rH over 4 weeks. Five replicates were analysed after 1 day and after 1, 2, 3 and 4 weeks. Freshly prepared samples were used as comparators to assess the stability of the ARV drugs. Circles indicate the calculated mean percentage stability and the error bars are standard deviations. Nevirapine and efavirenz concentrations were stable during 4 weeks of incubation at 40°C and 75% rH. Maximal deviation from freshly prepared samples was <5%, which is smaller than the accepted analytical error of ±15%. In contrast, lopinavir DBS concentrations decreased time-dependently by −14% after 2 weeks and by −25% after 4 weeks of incubation. Hence, the observed difference between DBS and plasma concentrations in the field study can be explained by degradation of lopinavir in DBS samples under the climatic conditions encountered in Tanzania. Discussion TDM of ARV has been shown to improve treatment outcome by preventing toxicity and insufficient viral suppression due to inadequate drug levels. However, TDM is frequently not implemented in resource-poor regions due to the shortage of infrastructure. Our field study shows that automated DBS analysis alone (e.g. for nevirapine) or in combination with an algorithm correcting for blood/plasma partitioning (e.g. for efavirenz) can successfully be applied in such a setting. Automation of the extraction process is central to establish DBS-based TDM in medical laboratories, as it improves robustness, sensitivity, efficiency and thus overall cost-effectiveness of measurements.28 To date, several techniques for the automation of DBS samples have been introduced,10,29–32 but only a few have so far been used to quantify concentrations in clinical patient samples.28,33 With our setup we demonstrated that a large number of DBS study samples can be analysed in a fully automated way. Accuracy and precision data show that the method was reliable (Table S1) and time-efficient, considering the analysis time of ∼4 min per sample. Hence, this type of automated DBS analysis holds promise to further DBS sampling for TDM. However, as serum or plasma and not venous or capillary blood is typically used for TDM analysis, a comparison of the two matrices is necessary. Factors affecting suitability of a matrix for TDM are the unbound drug fraction in plasma and blood, the haematocrit and the distribution of the analyte into RBCs. According to Emmons and Rowland34,35 the blood/plasma concentration ratio is a useful metric to select the appropriate matrix. Blood and plasma concentrations of nevirapine agreed strongly (ratio of 1.06), indicating that nevirapine distributes almost equally into the cellular and plasma compartments of the blood (KRBC/plasma 1.15). An in vivo study using radioactive-labelled nevirapine also showed an even distribution of nevirapine between human plasma and whole blood.36 Therefore, either plasma or whole blood could be used for TDM. In contrast, efavirenz and lopinavir are highly bound to plasma proteins;37–39 thus, distributing only partially into the cellular compartment resulting in low blood/plasma ratios of 0.62 and 0.57, respectively. To our knowledge, no data are available in the literature about the blood/plasma ratios of efavirenz and lopinavir. However, observed DBS/plasma ratios (efavirenz 0.6; lopinavir 0.51) were similar to blood/plasma ratios and importantly these data were comparable with published DBS/plasma ratios for efavirenz.20,40,41 At this low ratio, blood cells basically act as diluent and hence (under the assumption of constant protein binding), the haematocrit value primarily alters the proportion between blood and plasma concentrations. Thus, according to the decision tree proposed by Emmons and Rowland,35 blood and plasma can equally be used for all three ARV drugs assuming constancy of the parameters, e.g. haematocrit, the unbound drug fraction and the blood cell partitioning. However, the haematocrit values in our study population ranged between 10% and 50%. Consequently, concentrations measured in blood or DBS have to be adjusted by the haematocrit to allow reasonable prediction of the plasma concentration. We applied two different correction algorithms; both incorporated the haematocrit, as well as either the RBC partitioning or the plasma protein binding of the drugs.20,22,42 As expected, the agreement between plasma and blood concentrations was good for nevirapine but insufficient for efavirenz and lopinavir. Both methods of correction considerably improved agreements with only few samples exhibiting a bias >20%. Blood and DBS concentrations were largely similar, except that nevirapine concentrations in DBS samples tend to be ∼10% higher than in whole blood samples. Therefore, the agreement between blood and plasma was closer than for DBS and plasma. However, interpretation is difficult as the bias still lies within the generally accepted accuracy of 85%–115% for bioanalytical methods.23 Moreover, in the field study, which included a larger number of samples, almost no bias was detected between nevirapine concentrations determined in plasma and DBS samples. Similar results have been observed by Kromdijk and colleagues.20 Our data also showed that neither the haematocrit nor the protein binding altered the agreement between DBS and plasma, as the applied corrections had no impact. For nevirapine, the outcome of bridging and the field study was similar; thus, prolonged storage and shipment did not affect nevirapine analysis in DBS samples. This result is in line with our stability recorded under accelerated storage conditions and data published in the literature.43,44 The DBS and plasma comparison of efavirenz and lopinavir samples collected within the bridging study was very promising. The mean bias was <10% after correction by the haematocrit and protein binding or KRBC/plasma while almost none of the samples was beyond ±20%. The agreement between plasma and DBS samples collected within the field study was acceptable for efavirenz but not for lopinavir. The efavirenz concentration in DBS was on average 52% lower than in plasma, which is in line with studies from Kromdijk et al.20 and Amara et al.,40 who found a 39.8% and 41.9% decrease, respectively. Only a small negative bias of −10% was observed after correcting the DBS concentration by the haematocrit and protein binding and three-quarters of all samples exhibited a bias of <20%. This result further strengthens the usefulness of this algorithm to correlate DBS with plasma data as proposed by Li and Tse42 and Kromdijk et al.20 The small negative bias might be due to some decay of efavirenz; however, neither our stability experiments under accelerated storage conditions nor published data indicate relevant instability.20,40,41,43 In the case of lopinavir, the agreement between plasma and DBS samples collected in the field study was surprisingly poor, considering the promising data obtained in the bridging study. Overall, despite the correction, concentrations measured in DBS samples were ∼30% lower than in plasma, whereas the bias observed in the bridging study was negligible. Only a few studies about lopinavir analysis in DBS have been published so far,45–48 but, to the best of our knowledge, no data are available comparing concentrations in DBS and plasma samples. Our data indicate that fresh DBS samples can be used for TDM of lopinavir; however, prolonged storage and sample shipment affect the quality of DBS samples and lead to unreliable results. This is most likely due to degradation of lopinavir in DBS samples and fits our observations of lopinavir degradation under accelerated storage conditions. Lopinavir is generally considered to be stable at room temperature and after heat exposure at 58°C for 35 min.43,46,49–51 However, one study shows that the lopinavir concentration decreases by ∼13% in DBS kept at 20°C in desiccators for 2 years, which supports our assumption of lopinavir decaying in DBS over time.45 Our study shows that concentrations determined in whole blood are highly correlated with plasma concentrations. In the case of efavirenz and lopinavir, blood samples need to be corrected by the haematocrit and the cellular fraction of the drugs. Using the same correction algorithms, a good correlation between DBS and plasma concentrations of samples collected within the bridging study was reached. Agreement between plasma and DBS concentrations was also convincing for nevirapine and efavirenz for samples collected under field conditions in Tanzania. However, lopinavir was not stable in DBS under the conditions of this study, an aspect that should be considered when planning future studies. Importantly, the analysis of adjusted DBS and plasma concentrations lead to the same clinical interpretation of most TDM samples. Overall, automation facilitated DBS measurements by increasing method sensitivity and by decreasing the overall workload; thus, enhancing attractiveness of the DBS technique for TDM, pharmacokinetic studies and assessment of treatment adherence. However, before DBS can be implemented in low- and middle-income countries, obstacles, e.g. cost of sampling and analysis, have to be overcome. Acknowledgements We thank Beatrice Vetter and the clinical staff of the Chronic Diseases Clinic of Ifakara (Tanzania) for assistance with the study as well as all the patients that participated in the study. Funding S. K. was supported by a grant of the Swiss National Science Foundation (SNF 31003A_156270). The Chronic Diseases Clinic of Ifakara is funded by the Ministry of Health and Social Welfare of Tanzania, the Swiss Tropical and Public Health Institute, The Ifakara Health Institute, USAID through its local implementer TUNAJALI-Deloitte and the Government of the Canton of Basel. Transparency declarations S. G. is an employee of CAMAG (Muttenz, Switzerland). The remaining authors have none to declare. Author contributions U. D. and B. B. conducted analyses, wrote the first draft of the manuscript and were supervised by M. H. and S. K. M. D. was involved in the sample analyses within the bridging study. S. G. provided technical support for the automated bioanalyses. S. E. with oversight from M. B. and M. H. established the study design. S. E., A. N., D. M. and E. L. were involved with leading field data collection and ensuring the quality of data. All authors contributed to the writing of the paper and have read and approved the final manuscript. Supplementary data Additional Methods information, Table S1 and Figure S1 are available as Supplementary data at JAC Online. References 1 Gogtay NJ , Kshirsagar NA , Dalvi SS. Therapeutic drug monitoring in a developing country: an overview . Br J Clin Pharmacol 2001 ; 52 Suppl 1: 103S – 8S . Google Scholar Crossref Search ADS PubMed 2 Nwobodo N. Therapeutic drug monitoring in a developing nation: a clinical guide . JRSM Open 2014 ; 5 : 2054270414531121. Google Scholar Crossref Search ADS PubMed 3 Hosseinipour MC , Schechter M. Monitoring antiretroviral therapy in resource-limited settings: balancing clinical care, technology, and human resources . Curr HIV/AIDS Rep 2010 ; 7 : 168 – 74 . Google Scholar Crossref Search ADS PubMed 4 Koenig SP , Kuritzkes DR , Hirsch MS et al. Monitoring HIV treatment in developing countries . BMJ 2006 ; 332 : 602 – 4 . Google Scholar Crossref Search ADS PubMed 5 ter Heine R , Beijnen JH , Huitema AD. Bioanalytical issues in patient-friendly sampling methods for therapeutic drug monitoring: focus on antiretroviral drugs . Bioanalysis 2009 ; 1 : 1329 – 38 . Google Scholar Crossref Search ADS PubMed 6 Smit PW , Elliott I , Peeling RW et al. An overview of the clinical use of filter paper in the diagnosis of tropical diseases . Am J Trop Med Hyg 2014 ; 90 : 195 – 210 . Google Scholar Crossref Search ADS PubMed 7 Pandya HC , Spooner N , Mulla H. Dried blood spots, pharmacokinetic studies and better medicines for children . Bioanalysis 2011 ; 3 : 779 – 86 . Google Scholar Crossref Search ADS PubMed 8 Edelbroek PM , van der Heijden J , Stolk LM. Dried blood spot methods in therapeutic drug monitoring: methods, assays, and pitfalls . Ther Drug Monit 2009 ; 31 : 327 – 36 . Google Scholar Crossref Search ADS PubMed 9 Guthrie R , Susi A. A simple phenylalanine method for detecting phenylketonuria in large populations of newborn infants . Pediatrics 1963 ; 32 : 338 – 43 . Google Scholar PubMed 10 Abu-Rabie P , Spooner N , Chowdhry BZ et al. DBS direct elution: optimizing performance in high-throughput quantitative LC-MS/MS analysis . Bioanalysis 2015 ; 7 : 2003 – 17 . Google Scholar Crossref Search ADS PubMed 11 Duthaler U , Berger B , Erb S et al. Automated high throughput analysis of antiretroviral drugs in dried blood spots . J Mass Spectrom 2017 ; 52 : 534 – 42 . Google Scholar Crossref Search ADS PubMed 12 Kappelhoff BS , van Leth F , Robinson PA et al. Are adverse events of nevirapine and efavirenz related to plasma concentrations? Antiviral Ther 2005 ; 10 : 489 – 98 . 13 Wood E , Hogg RS , Yip B et al. Effect of medication adherence on survival of HIV-infected adults who start highly active antiretroviral therapy when the CD4+ cell count is 0.200 to 0.350 × 109 cells/L . Ann Intern Med 2003 ; 139 : 810 – 16 . Google Scholar Crossref Search ADS PubMed 14 Hogg RS , Heath K , Bangsberg D et al. Intermittent use of triple-combination therapy is predictive of mortality at baseline and after 1 year of follow-up . AIDS 2002 ; 16 : 1051 – 8 . Google Scholar Crossref Search ADS PubMed 15 Paterson DL , Swindells S , Mohr J et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection . Ann Intern Med 2000 ; 133 : 21 – 30 . Google Scholar Crossref Search ADS PubMed 16 Burkhart PV , Sabaté E. Adherence to long-term therapies: evidence for action . J Nurs Scholarsh 2003 ; 35 : 207 . Google Scholar PubMed 17 World Heath Organization . HIV/AIDS—Fact Sheet No. 360 (updated July 2017 ). http://www.who.int/mediacentre/factsheets/fs360/en/. 18 Erb S , Letang S , Glass T et al. Health care provider communication training in rural Tanzania empowers HIV-infected patients on antiretroviral therapy to discuss adherence problems . HIV Med 2017 ; 18 : 623 – 34 . Google Scholar Crossref Search ADS PubMed 19 FDA . Guidance for Industry Bioanalytical Method Validation. http://www.fda.gov/cder/guidance/index.htm. 20 Kromdijk W , Mulder JW , Rosing H et al. Use of dried blood spots for the determination of plasma concentrations of nevirapine and efavirenz . J Antimicrob Chemother 2012 ; 67 : 1211 – 16 . Google Scholar Crossref Search ADS PubMed 21 Boffito M , Back DJ , Blaschke TF et al. Protein binding in antiretroviral therapies . AIDS Res Hum Retroviruses 2003 ; 19 : 825 – 35 . Google Scholar Crossref Search ADS PubMed 22 Jager NG , Rosing H , Schellens JH et al. Use of dried blood spots for the determination of serum concentrations of tamoxifen and endoxifen . Breast Cancer Res Treat 2014 ; 146 : 137 – 44 . Google Scholar Crossref Search ADS PubMed 23 EMEA . Guideline on Bioanalytical Method Validation 2011; 2016 (3 January). http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2011/08/WC500109686.pdf. 24 Briggs RJ , Nicholson R , Vazvaei F et al. Method transfer, partial validation, and cross validation: recommendations for best practices and harmonization from the global bioanalysis consortium harmonization team . AAPS J 2014 ; 16 : 1143 – 8 . Google Scholar Crossref Search ADS PubMed 25 Csajka C , Marzolini C , Fattinger K et al. Population pharmacokinetics and effects of efavirenz in patients with human immunodeficiency virus infection . Clin Pharmacol Ther 2003 ; 73 : 20 – 30 . Google Scholar Crossref Search ADS PubMed 26 Guidi M , Arab-Alameddine M , Rotger M et al. Dosage optimization of treatments using population pharmacokinetic modeling and simulation . Chimia 2012 ; 66 : 291 – 5 . Google Scholar Crossref Search ADS PubMed 27 Lubomirov R , di Iulio J , Fayet A et al. ADME pharmacogenetics: investigation of the pharmacokinetics of the antiretroviral agent lopinavir coformulated with ritonavir . Pharmacogenet Genomics 2010 ; 20 : 217 – 30 . Google Scholar PubMed 28 Fingerhut R , Silva Polanco ML , Silva Arevalo Gde J et al. First experience with a fully automated extraction system for simultaneous on-line direct tandem mass spectrometric analysis of amino acids and (acyl-)carnitines in a newborn screening setting . Rapid Commun Mass Spectrom 2014 ; 28 : 965 – 73 . Google Scholar Crossref Search ADS PubMed 29 Verplaetse R , Henion J. Quantitative determination of opioids in whole blood using fully automated dried blood spot desorption coupled to on-line SPE-LC-MS/MS . Drug Test Anal 2016 ; 8 : 30 – 8 . Google Scholar Crossref Search ADS PubMed 30 Oliveira RV , Henion J , Wickremsinhe E. Fully-automated approach for online dried blood spot extraction and bioanalysis by two-dimensional-liquid chromatography coupled with high-resolution quadrupole time-of-flight mass spectrometry . Anal Chem 2014 ; 86 : 1246 – 53 . Google Scholar Crossref Search ADS PubMed 31 Ganz N , Singrasa M , Nicolas L et al. Development and validation of a fully automated online human dried blood spot analysis of bosentan and its metabolites using the sample card and prep DBS System . J Chromatogr B Analyt Technol Biomed Life Sci 2012 ; 885–886 : 50 – 60 . Google Scholar Crossref Search ADS PubMed 32 Abu-Rabie P , Spooner N. Direct quantitative bioanalysis of drugs in dried blood spot samples using a thin-layer chromatography mass spectrometer interface . Anal Chem 2009 ; 81 : 10275 – 84 . Google Scholar Crossref Search ADS PubMed 33 Tretzel L , Thomas A , Piper T et al. Fully automated determination of nicotine and its major metabolites in whole blood by means of a DBS online-SPE LC-HR-MS/MS approach for sports drug testing . J Pharm Biomed Anal 2016 ; 123 : 132 – 40 . Google Scholar Crossref Search ADS PubMed 34 Rowland M , Emmons GT. Use of dried blood spots in drug development: pharmacokinetic considerations . AAPS J 2010 ; 12 : 290 – 3 . Google Scholar Crossref Search ADS PubMed 35 Emmons G , Rowland M. Pharmacokinetic considerations as to when to use dried blood spot sampling . Bioanalysis 2010 ; 2 : 1791 – 6 . Google Scholar Crossref Search ADS PubMed 36 Riska P , Lamson M , MacGregor T et al. Disposition and biotransformation of the antiretroviral drug nevirapine in humans . Drug Metab Dispos 1999 ; 27 : 895 – 901 . Google Scholar PubMed 37 Illamola SM , Labat L , Benaboud S et al. Determination of total and unbound concentrations of lopinavir in plasma using liquid chromatography-tandem mass spectrometry and ultrafiltration methods . J Chromatogr B Analyt Technol Biomed Life Sci 2014 ; 965 : 216 – 23 . Google Scholar Crossref Search ADS PubMed 38 Almond LM , Hoggard PG , Edirisinghe D et al. Intracellular and plasma pharmacokinetics of efavirenz in HIV-infected individuals . J Antimicrob Chemother 2005 ; 56 : 738 – 44 . Google Scholar Crossref Search ADS PubMed 39 Dumond JB , Rigdon J , Mollan K et al. Brief report: significant decreases in both total and unbound lopinavir and amprenavir exposures during coadministration: ACTG Protocol A5143/A5147s results . J Acq Immune Defic Syndr 2015 ; 70 : 510 – 14 . Google Scholar Crossref Search ADS 40 Amara AB , Else LJ , Tjia J et al. A validated method for quantification of efavirenz in dried blood spots using high-performance liquid chromatography-mass spectrometry . Ther Drug Monit 2015 ; 37 : 220 – 8 . Google Scholar Crossref Search ADS PubMed 41 Hoffman JT , Rossi SS , Espina-Quinto R et al. Determination of efavirenz in human dried blood spots by reversed-phase high-performance liquid chromatography with UV detection . Ther Drug Monit 2013 ; 35 : 203 – 8 . Google Scholar Crossref Search ADS PubMed 42 Li W , Tse FL. Dried blood spot sampling in combination with LC-MS/MS for quantitative analysis of small molecules . Biomed Chromatogr 2010 ; 24 : 49 – 65 . Google Scholar Crossref Search ADS PubMed 43 ter Heine R , Rosing H , van Gorp EC et al. Quantification of protease inhibitors and non-nucleoside reverse transcriptase inhibitors in dried blood spots by liquid chromatography-triple quadrupole mass spectrometry . J Chromatogr B Analyt Technol Biomed Life Sci 2008 ; 867 : 205 – 12 . Google Scholar Crossref Search ADS PubMed 44 Olagunju A , Amara A , Waitt C et al. Validation and clinical application of a method to quantify nevirapine in dried blood spots and dried breast-milk spots . J Antimicrob Chemother 2015 ; 70 : 2816 – 22 . Google Scholar Crossref Search ADS PubMed 45 Meesters RJ , Hooff GP , Gruters R et al. Incurred sample reanalysis comparison of dried blood spots and plasma samples on the measurement of lopinavir in clinical samples . Bioanalysis 2012 ; 4 : 237 – 40 . Google Scholar Crossref Search ADS PubMed 46 Meesters RJ , van Kampen JJ , Reedijk ML et al. Ultrafast and high-throughput mass spectrometric assay for therapeutic drug monitoring of antiretroviral drugs in pediatric HIV-1 infection applying dried blood spots . Anal Bioanal Chem 2010 ; 398 : 319 – 28 . Google Scholar Crossref Search ADS PubMed 47 Koal T , Burhenne H , Römling R et al. Quantification of antiretroviral drugs in dried blood spot samples by means of liquid chromatography/tandem mass spectrometry . Rapid Commun Mass Spectrom 2005 ; 19 : 2995 – 3001 . Google Scholar Crossref Search ADS PubMed 48 ter Heine R , Davids M , Rosing H et al. Quantification of HIV protease inhibitors and non-nucleoside reverse transcriptase inhibitors in peripheral blood mononuclear cell lysate using liquid chromatography coupled with tandem mass spectrometry . J Chromatogr B Analyt Technol Biomed Life Sci 2009 ; 877 : 575 – 80 . Google Scholar Crossref Search ADS PubMed 49 Watanabe K , Varesio E , Hopfgartner G. Parallel ultra high pressure liquid chromatography-mass spectrometry for the quantification of HIV protease inhibitors using dried spot sample collection format . J Chromatogr B Analyt Technol Biomed Life Sci 2014 ; 965 : 244 – 53 . Google Scholar Crossref Search ADS PubMed 50 D’Avolio A , Simiele M , Siccardi M et al. HPLC-MS method for the quantification of nine anti-HIV drugs from dry plasma spot on glass filter and their long term stability in different conditions . J Pharm Biomed Anal 2010 ; 52 : 774 – 80 . Google Scholar Crossref Search ADS PubMed 51 Verbesselt R , Van Wijngaerden E , de Hoon J. Simultaneous determination of 8 HIV protease inhibitors in human plasma by isocratic high-performance liquid chromatography with combined use of UV and fluorescence detection: amprenavir, indinavir, atazanavir, ritonavir, lopinavir, saquinavir, nelfinavir and M8-nelfinavir metabolite . J Chromatogr B Analyt Technol Biomed Life Sci 2007 ; 845 : 51 – 60 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Using dried blood spots to facilitate therapeutic drug monitoring of antiretroviral drugs in resource-poor regions
JO - Journal of Antimicrobial Chemotherapy
DO - 10.1093/jac/dky254
DA - 2018-10-01
UR - https://www.deepdyve.com/lp/oxford-university-press/using-dried-blood-spots-to-facilitate-therapeutic-drug-monitoring-of-rpqBhZBh8i
SP - 2729
VL - 73
IS - 10
DP - DeepDyve
ER -