Abstract
Objectives
Regimens containing protease inhibitors (PI) are less commonly used in developing countries due to high cost and less availability. We evaluated characteristics of patients initiating PI-based therapy according to previous antiretroviral (ARV) exposure; factors associated with initiating a PI-containing regimen using newer versus older PIs, and proportion of patients with detectable viral loads (VL) after initiating a PI-based regimen.
Methods
This analysis includes all patients who have initiated a PI-based regimen. ARV exposure was categorised: naïve (no previous ARV), 1st, 2nd, ≥ 3rd switches; a switch was defined as starting or stopping any drug in a regimen. Newer PIs were defined as those approved by the US FDA after 1 January 2000. Detectable VL at 12 months was defined as VL ≥ 400 copies/mL. Characteristics at PI initiation were evaluated. Logistic regression was used to determine factors associated with initiating a newer PI and detectable VL at 12 months after PI initiation.
Results
1106 patients initiated PI-based therapy; of these, 618 (56%) were naïve patients. Overall, 22% (176) of patients had detectable VL at 12 months following the PI initiation. Being from a high income country (vs. mid/low income, OR = 1.80, p = 0.034) were more likely to be associated with detectable VL.
Conclusion
The use of PIs in this cohort is dictated by accessibility and affordability issues particularly for the newer PIs. Short-term virological outcomes following PI-therapy in our cohort were good, and were associated with CD4 count at time of initiation.
Keywords: HIV, HAART, Disease progression, Protease inhibitor
Introduction
Antiretroviral therapy (ART) has been associated with improved morbidity and mortality among persons with human immune deficiency Virus (HIV) infection (Palella et al., 2006). Currently 3 million patients have started taking first line ART in resource-limited settings (RLS). However, a significant proportion of patients fail first line ART for various reasons, particularly non-adherence (Gardner et al., 2008; Ramadhani et al., 2007). As treatment cohorts mature, the need for second line protease inhibitor (PI) based regimens will increase.
There is very limited information on the use of PIs in RLS (Pujades-Rodriguez et al., 2008). Protease inhibitors are more expensive than first line NNRTI based regimen, thus limiting their use. Thus the characteristics of patients initiating PIs may be dictated more by access and affordability in RLS. Use of PIs has also been associated with toxicities, particularly metabolic. We explored characteristics of patients initiating PI based regimens, factors associated with use of newer versus older PIs and effectiveness of PI-based regimes in RLS amongst a large cohort of patients from the Asia-Pacific region.
Methods
The Cohort
TREAT Asia HIV Observational Database (TAHOD) is a collaborative observational cohort study, currently involving 17 participating clinical sites across the Asia-Pacific region. Each site has recruited approximately 200 patients and collected demographic, and clinical data such as antiretroviral treatment, AIDS diagnosis, CD4 counts, viral load and other clinical and laboratory data. The nature and characteristics of the cohort have been described elsewhere (Zhou and Kumarasamy, 2005; Zhou et al., 2005). Ethics approval for the study was obtained from the University of New South Wales Ethics committee and the local ethics committee of each participating TAHOD site.
Inclusion Criteria
Data from PI initiation up to September 2007 were included in these analyses.
For the analysis of predictors for and effectiveness of PI-based therapy, all patients who initiated a PI were included and were categorised by previous ART exposure; naïve, first switch, second switch, third or greater switch; a switch was defined as starting or stopping any component of the ART regimen. For the analysis of factors associated with using older versus newer PIs, patients were included if they initiated their first PI therapy after 2000. Newer PIs were defined as those that were approved by the United States Food and Drug Administration (FDA) after 1 January 2000; these include lopinavir/ritonavir, atazanavir, amprenavir, fos-amprenavir, tipranavir and darunavir.
Measurements
The measurement closest to the date prior to PI initiation was used for baseline CD4 count, viral load and lipid laboratory results. Measurements were considered within the time window of 90 days before the date of starting the PI and up to 14 days after initiation. For the 6 and 12 month nominal visit times, a time window within +/− 3 months of the nominal visit time was adopted.
Statistics
Descriptive statistics were used to summarise characteristics of patients at PI initiation. Variables included country income (low, lower-middle, upper-middle and high income) as defined by the World Bank, generic or patent PIs, boosted RTV versus non-boosted RTV, older versus newer PIs, CD4 count, viral load and duration from HIV diagnosis to PI initiation.
Logistic regression was used to determine factors associated with initiating PI therapy using newer versus older PIs. Variables included in this analysis were prior ART exposure (naïve versus 1st switch, 2nd switch and 3rd + switch), country income (high versus other income), RTV dose, generic versus patent PI, CD4 and viral load at PI initiation, and duration from HIV diagnosis to PI initiation.
Characteristics of patients with detectable and undetectable viral load (</>400 copies/ml) at 12 months were summarised at baseline, 6 months and at 12 months. Variables summarised included CD4, viral load, cholesterol, HDL-cholesterol, triglycerides and glucose. Logistic regression was used to determine factors associated with detectable viral load at 12 months. Patients who started PI and had viral load measurement at 12 months were included in this analysis. Factors included in the univariate models were prior ART exposure, country income, generic versus patent PI, CD4 and viral load at PI initiation, clinical adverse events at 12 month, duration from HIV diagnosis to PI initiation and duration of prior ART.
Multivariate analyses were performed using a forward stepwise approach. Variables with p< 0.10 in univariate analyses were considered for inclusion in multivariate analyses. All statistical analyses were performed using STATA 10.0 software, and all p-values were two-sided. A level of significance at 0.05 was used throughout these analyses.
Results
Patient Characteristics
Of 3,821 patients in the TAHOD database, 1,102 initiated PI containing regimens; 612 of these were naive at PI initiation. Characteristics of the patients are described in Table 1. Six hundred forty-seven patients (58.7%) were from high income countries. The majority of patients (63%) used patented PIs rather than generic. Median (IQR) of CD4 cells count at baseline was 134 (38.5 – 242.5) cells/μL and median (IQR) of log10 HIV-RNA was 4.7 (4.0 – 5.3) copies/mL. Median time to first PI use from first HIV diagnosis was 1.3 (0.2 – 3.7) years.
Table 1.
Baseline characteristics of patients starting PIs.
| ARV naïve (n = 612) | 1st switch (n = 207) | 2nd switch (n = 149) | >= 3rd switch (n = 135) | Total (n = 1102) | |
|---|---|---|---|---|---|
| Country income | |||||
| - High income | 429 (70) | 106 (51.2) | 64 (43) | 48 (35.6) | 647(55.44) |
| - Low income | 9 (1.5) | 15 (7.2) | 17 (11.4) | 14 (10.4) | 56(18.78) |
| - Lower-middle income | 131 (21.4) | 35 (17.0) | 37 (24.8) | 33 (24.4) | 236(13.52) |
| - Upper-middle income | 42 (7.0) | 51 (24.6) | 31 (20.8) | 40 (29.6) | 164(12.25) |
| Generic/Patent drugs | |||||
| - Generic | 37 (6.3) | 40 (18.7) | 22 (15.9) | 26 (19.4) | 125 (11.3) |
| - Patent | 383 (62.2) | 121 (58.9) | 101 (68.3) | 89 (65.6) | 694 (63) |
| - Don’t know | 10 (1.6) | 10 (4.8) | 4 (2.8) | 12 (9.0) | 36 (3.3) |
| - Missing | 181 (29.9) | 36 (17.7) | 22 (13.1) | 8 (6.0) | 247 (22.4) |
| RTV dose (n = 399) | |||||
| - full dose | 11 (7.3) | 6 (8.4) | 3 (4.4) | 3 (4.2) | 23 (6.3) |
| - low dose | 140 (92.7) | 77 (91.6) | 71 (95.7) | 94 (95.8) | 382 (93.7) |
| FDA approval | |||||
| - older (before 2000) | 333 (55.3) | 169 (82.3) | 125 (83.5) | 112 (82.8) | 739 (67.1) |
| - newer (from 2000) | 278 (44.7) | 38 (17.2) | 24 (16.6) | 23 (16.4) | 363 (32.9) |
| CD4 cells/mL | |||||
| - Mean (SD) | 193.6 (156.3) | 138.8 (152.5) | 144.1 (168.2) | 134.6 (123.3) | 166.4(154.5) |
| - Median (IQR) | 180 (60 – 71) | 79 (29 – 191) | 92.5 (37 – 03) | 96 (37 – 210) | 134 (38.5 – 242.5) |
| Log10 HIV-RNA, log10 copies/mL | |||||
| - Mean (SD) | 4.7 (1.0) | 4.3 (1.1) | 4.2 (0.9) | 4.0 (1.1) | 4.5 (1.1) |
| - Median (IQR) | 4.9 (4.2 – 5.5) | 4.5 (3.7 – 5.1) | 4.4 (3.7 – 4.9) | 4.1 (3.4 – .8) | 4.7 (4.0 – 5.3) |
| Years since HIV diagnosis to PI used | |||||
| - Mean (SD) | 1.6 (2.5) | 2.6 (3.1) | 3.9 (3.1) | 4.6 (3.0) | 2.4 (3.0) |
| - Median (IQR) | 0.4 (0.1 – 2.0) | 1.6 (0.3 – 3.5) | 3.0 (1.3 – 5.8) | 4.1 (2.6 – 6.0) | 1.3 (0.2 – 3.7) |
NB: PI naïve is patients used a PI as their first treatment,
1st switch is when patients first included PI in their second regimen,
2nd switch is when patients first included a PI in their 3rd regimen and
3rd and more switch is when patients first included a PI after their 4th regimen.
Indinavir was the most commonly used PI (48%) and ritonavir-boosted regimens were used by 53.6% (Table 2). Treatment failure was the main reason for experienced patients to switch to a PI-based regimen (26%), while adverse events were reported as the reason in 15.7% of patients who switched to a PI based regimen (Table 3).
Table 2.
Distribution of first PI drugs used in TAHOD.
| PI | Frequency |
|---|---|
| Atazanavir/r | 64 |
| Lopinavir/r | 216 |
| Indinavir/r | 218 |
| SQV/r | 69 |
| Tipranavir/r | 21 |
| Fosamprenvir/r | 5 |
| Danunavir/r | 2 |
| Atazanavir | 53 |
| Indinavir | 317 |
| Nelfinavir | 100 |
| Ritonavir (full dose) | 7 |
| SQV | 34 |
| Fosamprenavir | 1 |
| Lopinavir | 2 |
patients may use more than 1 PI
Table 3.
Reported reasons for stopping the ARV regimen that lead to first PI use.
| Reasons | n (%) |
|---|---|
| Treatment failure | 130 (26.5) |
| Clinical progression/hospitalisation | 33 (6.7) |
| Patient’s decision/request | 22 (4.5) |
| Compliance difficulties | 36 (7.3) |
| Drug interaction | 3 (0.6) |
| Adverse events | 77 (15.7) |
| Other | 23 (4.7) |
| No information | 95 (19.3) |
| drug change | 28 (5.7) |
| enrol to a new study | 32 (6.5) |
| financial | 2 (0.4) |
| physician decision | 5 (1.0) |
| resistance | 5 (1.0) |
Predictive Factors for Patients using a Newer PI (FDA approval after 2000)
Table 4 shows factors associated with patients using newer PIs. In the univariate analyses, factors related to the use of newer PI were: naive at PI initiation (p < 0.001), high income countries (p < 0.001), use of a patent PI (p < 0.001), CD4 200 – 350 cells/μL at PI initiation (p = 0.01), viral load more than 50,000 copies/mL at PI initiation (p = 0.002) and longer duration from HIV diagnosis to PI initiation (p = 0.03).
Table 4.
Predictive factors for patients using newer PI (FDA approved after 2000).
| Older PI | Newer PI | Univariate results | Multivariate results | |||
|---|---|---|---|---|---|---|
| N(%) | N(%) | OR (95% CI) | P | OR (95% CI) | P | |
| Total number | 652 | 387 | ||||
| ARV regimen in which PI first used | < 0.001 | < 0.001 | ||||
| - naïve | 225 (34) | 278 (72) | 1.00 | … | 1.00 | … |
| - 1st switch | 137 (21) | 40 (10) | 0.24 (0.16 – 0.35) | < 0.001 | 0.26 (0.16 – 0.42) | < 0.001 |
| - 2nd switch | 122 (19) | 26 (7) | 0.17 (0.11 – 0.27) | < 0.001 | 0.15 (0.09 – 0.25) | < 0.001 |
| - 3rd and more switch | 168 (26) | 43 (11) | 0.21 (0.14 – 0.30) | < 0.001 | 0.18 (0.12 – 0.27) | < 0.001 |
| Country income | ||||||
| - mid/low | 327 (50) | 100 (26) | 1.00 | … | 1.00 | … |
| - high income | 325 (50) | 287 (74) | 2.88 (2.19 – 3.80) | < 0.001 | 1.57 (1.10 – 2.23) | 0.012 |
| Generic v patent | ||||||
| - Generic | 112 (23) | 4 (1) | 1.00 | 1.00 | … | |
| - Patent | 352 (71) | 309 (99) | 24.58 (8.96 – 67.41) | < 0.001 | 14.54 (5.09 – 41.54) | < 0.001 |
| CD4 at PI initiation, cells/mL | 0.001 | |||||
| - < 200 | 256 (54) | 109 (50) | 1.00 | … | ||
| - 200 – 350 | 129 (28) | 83 (38) | 1.51 (1.06 – 2.16) | 0.023 | ||
| - > 350 | 83 (18) | 26 (12) | 0.74 (0.45– 1.21) | 0.224 | ||
| RNA at PI initiation, copies/mL | ||||||
| - < 50000 | 231 (64) | 100 (51) | 1.00 | … | ||
| - 50000 + | 129 (36) | 97 (49) | 1.74 (1.22 – 2.47) | 0.002 | ||
| Years since HIV diagnosis to PI used | ||||||
| - Mean (SD) | 3.2 (3.2) | 2.8 (3.3) | 0.96 (0.92 – 0.998) | 0.034 |
NB: Patients included in this analysis if they initiated their first PI therapy after 2000.
Multivariate analysis showed that patients with prior ART exposure were less likely to use a newer PI compared with naive patients (1st switch: OR 0.22 [95% confidence interval 0.13–0.38], p < 0.001; 2nd switch: OR 0.12 [0.07–0.20], p < 0.001; =3rd switch: OR 0.14 [0.08–0.22], p < 0.001). Patients from high income countries were more likely to use a newer PI (OR 1.67[1.18 – 2.38], p = 0.005), and those on patent PIs were more likely to use a newer PI (OR 18.64 [5.64 – 61.60], p < 0.001).
Characteristics of Patients at Time of PI Initiation
Table 5 summarises factors associated with having an undetectable VL at 12 months after starting PIs. Ninety per cent of patients had a detectable viral load at PI initiation and 74.3% achieved an undetectable viral load 12 months after starting PI-based therapy. Patients who had undetectable viral load at 12 months had a higher median CD4 count (162, range 50–264) compared to those with detectable viral loads at 12 months (96, range 31–182).
Table 5.
Characteristics of patients at baseline, 6 and 12 months after starting PI according to viral load status at 12 months following PI.
| Vl < 400copies/ml at 12 months (n =507) | VL >= 400 copies/ml at 12 months (n = 175) | |||||
|---|---|---|---|---|---|---|
| Baseline | 6 months | 12 months | Baseline | 6 months | 12 months | |
| CD4, cells/mL (n) | 441 | 452 | 496 | 142 | 145 | 168 |
| - Mean (SD) | 185.9 (157.3) | 308.5 (201.4) | 360.2 (209) | 144.3 (158.3) | 250.5 (195.2) | 283 (197.7) |
| - Median (IQR) | 162 (50 – 264) | 278 (149 – 418) | 326 (206 – 466.5) | 96.5 (31 – 182) | 198 (109 – 339) | 253.5 (135 – 374.5 |
| Changes in CD4, cells/μL (n) | 326 | 339 | 107 | 122 | ||
| - Mean (SD) | NA | 73.8 (179.8) | 130 (183.4) | NA | 16.5 (179.5) | 53.0 (164.6) |
| - Median (IQR) | NA | 89.5 (−17 – 167) | 131 (17 – 230) | NA | 17 (−91 – 122) | 39.5 (−43 – 147) |
| Detectable VL (VL > 400 copies/μL) (n) | 306 | 422 | 507 | 100 | 129 | 175 |
| - VL < 400 copies/mL | 77 (9.0) | 368 (78.0) | 507 (100.0) | 22 (8.0) | 42 (23.0) | 0 (0) |
| - VL ≥ 400 copies/mL | 229 (91.0) | 54 (22.0) | 0 (0) | 78 (92.0) | 87(77.0) | 175 (100) |
| Cholesterol, mmol/L (n) | 40 | 201 | 254 | 16 | 63 | 81 |
| - Mean (SD) | 1 (2.0) | 1.9 (3.0) | 1.8 (2.7) | 1.3 (2.5) | 2.4 (2.5) | 2.5 (2.7) |
| Changes in Cholesterol, mmol/L | 34 | 38 | 12 | 13 | ||
| - Mean (SD) | NA | 0.05 (0.1) | 0.2 (0.5) | NA | 0.06 (0.2) | −0.02 (0.6) |
| HDL-Cholesterol, mmol/L (n) | 8 | 61 | 81 | 1 | 11 | 12 |
| - Mean (SD) | 0.9 (0.6) | 1 (3) | 0.5 (0.7) | 1.3, n = 1 | 0.7 (0.7) | 0.5 (0.6) |
| Changes in HDL-Cholesterol, mmol/ | 5 | 8 | ||||
| - Mean (SD) | NA | −0.1 (0.3), | 0.04 (0.3) | NA | No Data | No Data |
| Triglycerides, mmol/L (n) | 12 | 152 | 205 | 8 | 48 | 59 |
| - Mean (SD) | 1.2 (1.2) | 1.6 (3.5) | 1.4 (3.2) | 1 (1.1) | 1.3 (1.2) | 1.8 (2.0) |
| Changes in Triglycerides, mmol/L (n) | 8 | 11 | 5 | 5 | ||
| - Mean (SD) | NA | 0.9 (0.8) | 2.3 (5.4) | NA | −0.1(0.3) | 0.9 (1.1) |
| Glucose, mmol/L (n) | 40 | 13 | ||||
| - Mean (SD) | 1.2 (2.3) | No Data | No Data | 1.2 (2.3) | No Data | No Data |
| Changes in Glucose, mmol/L (n) | ||||||
| - Mean (SD) | NA | No Data | No Data | NA | No Data | No Data |
Predictive Factors of Detectable Viral Loads at 12 Months
Table 6 shows predictive factors of detectable viral loads at 12 months. In univariate analyses, factors associated with detectable viral loads were; use of PI in a second or subsequent PI regimen (p = 0.003), and CD4 200 – 350 cells/μL at PI initiation (p < 0.031). In the multivariate analysis when adjusted for both the ARV regimen in which a PI was first used and CD4 at PI initiation only CD4 at PI initiation remained statistically significant to predict detectable viral load at 12 months (Table 6).
Table 6.
| VL < 400 copies/mL | VL > 400 copies/mL | Univariate results | ||
|---|---|---|---|---|
| N(%) | N(%) | OR (95% CI) | P | |
| Total number | 507 | 175 | ||
| ARV regimen in which PI first used | 0.003 | |||
| - naïve | 301 (59) | 82 (47) | 1.00 | … |
| - 1st switch | 95 (19) | 38 (22) | 1.46(0.94 – 2.30) | 0.093 |
| - 2nd switch | 62 (12) | 31 (17) | 1.84 (1.12 – 3.01) | 0.016 |
| - 3rd and more switch | 49 (10) | 24 (14) | 1.80 (1.04 – 3.10) | 0.035 |
| Country income | ||||
| - mid/low | 159 (31) | 56 (32) | 1.00 | … |
| - high income | 348 (69) | 119 (68) | 0.97 (0.67 – 1.40) | 0.875 |
| Generic v patent | ||||
| - Generic | 52 (10) | 23 (13) | 1.00 | … |
| - Patent | 323(64) | 109 (62) | 0.76 (0.45 – 1.30) | 0.323 |
| CD4 at PI initiation, cells/μL | 0.089 | |||
| - < 200 | 155 (45) | 68 (54) | 1.00 | … |
| - 200 – 350 | 124 (36) | 32 (25) | 0.59 (0.36 – 0.95) | 0.031 |
| - > 350 | 64 (19) | 26 (21) | 0.93 (0.54 – 1.56) | 0.779 |
| RNA at PI initiation, copies/mL | ||||
| - < 50000 | 186 (61) | 59 (59) | 1.00 | … |
| - 50000 + | 120 (39) | 41 (41) | 1.08 (0.68 – 1.71) | 0.394 |
| Clinical Adverse events at 12 months | ||||
| - No AE | 501 (99) | 172 (98) | 1.00 | … |
| - AE | 6 (1) | 3 (2) | 1.46 (0.36 – 5.89) | 0.598 |
| Years since HIV diagnosis to PI initiation | N = 503 | N = 175 | ||
| - Mean (SD) | 2.1 (2.8) | 2.3 (2.8) | 1.02 (0.96 – 1.08) | 0.504 |
The only variable remained statistically significant in the multivariate model is CD4 counts between 200 – 300 cells/μL at PI initiation
Patients included in this analysis if they started PI and had viral load detectable at 12 months.
As a sensitivity analysis, we also conducted separate analyses for naive and experienced patients. For naive patients, the only predictor of detectable viral load at 12 months was a lower CD4 at PI initiation and it was similar to overall results. For experienced patients, the predictor of detectable viral loads at 12 months was being from a high income country (data not shown).
Discussion
We have described here the experience of PI use in a large cohort of patients in the Asia-Pacific region. Of the total cohort initiating PIs, almost 50% used it in first line regimens. Using PI-based regimens for naïve patients was more common in high-income countries, and patients had access to ART earlier in the course of disease. More common use of patented PIs over generic PIs may be due to later availability of the generic PIs and patent protection in higher income countries. Use of generic PIs is expected to increase as these drugs become widely available. Currently, saquinavir, indinavir, ritonavir (including the heat stable formulation), lopinavir/ritonavir (heat stable formulation), and atazanavir are available in generic formulations. Indinavir (IDV) is the most commonly prescribed PI in this cohort, possibly because it was one of first PIs to become available in patented and generic versions. The effectiveness of IDV-based regimens has been documented in RLS. Studies have reported the effectiveness and lower toxicity rates in using a lower dose of indinavir, using the favourable pharmacokinetic effects of ritonavir boosting. (Duvivier et al., 2003; Mootsikapun et al., 2005). However, the use of indinavir may decline as more potent and less toxic alternative PIs become increasingly accessible.
Although newer PIs retain effectiveness in patients with some PI resistance (Hicks et al., 2006; Pellegrin et al., 2008), they are used as first line therapy and not just for second-line and salvage ART regimens in high income countries within TAHOD. However, when used for virologic failure, resistance testing would be useful to determine the potential effectiveness of these drugs. Studies indicate that darunavir is non-inferior to lopinavir/ritonavir in naive patients, and may even be superior in those with high baseline viral loads (Ortiz et al., 2008). Newer PIs also provide survival benefit as compared to use of older PIs (Crane et al., 2007). Lack of access to these PIs may have contributed to patients continuing on a failing PI based regimens or switches between older PIs. This results in poor virologic suppression rates. Improved access to these drugs is urgently needed to better optimize the use of PIs in naive and treatment experienced patients in RLS.
Treatment failure was the most common reason for switching to a PI based regimen among ART experienced patients. PI based regimens are robust and have been proposed to be effective second line regimens after failing first line NNRTI-based regimens (Gallant, 2007; Martinez-Cajas and Wainberg, 2008). This is especially true if failure has been identified early (by virologic rather than immunologic or clinical criteria). Patients in TAHOD who switched because of failure had lower CD4 counts (all <200/mm3) reflecting late identification and treatment of failure, largely due to limited access to viral load testing and problems with affordability of PI therapy. Adverse events to NNRTI based regimens also led to substitution with PIs. Using PIs is the only option in patients with serious adverse events associated with use of both the first generation NNRTIs (Nevirapine and Efavirenz).
In spite of the limitations discussed above. PI based regimens including atazanavir, lopinavir, indinavir and other PIs had excellent virologic suppression rates at 12 months post initiation. This indicates the robustness of the PI or PI/r that significantly contributes to the overall regimen effectiveness. This is especially true when the backbone nucleosides are usually recycled and chosen without performing resistance testing. Protease inhibitors are highly potent drugs and have even shown to be effective as monotherapy in certain situations (Pierone et al., 2006; Schechter and Nunes, 2007). However, the only factor associated with poor effectiveness was a low CD4 count prior to switch indicating that PI based regimens may have limited effectiveness if failure is identified late. The consequence of identifying failure late is the accumulation of resistance mutations that can lead to further intra-class cross resistance thus limiting future options (Agwu et al., 2008; Sungkanuparph et al., 2007). There may be variations in the evolution of resistance amongst various subtypes (Cavalcanti et al., 2007). This calls for improving on the current protocols for identifying failures in ARV programs, particularly improved access to viral load testing. Availability of routine viral load testing may have contributed to more failures being reported from high income countries on PI based regimens in this cohort.
Apart from being an observational design, there are numerous limitations that need to be kept in mind. Since this cohort was established six years ago, the choice of PIs seems to reflect the availability at that point of time. As the cohort matures, use of newer PIs may increase. The sites contributing to data are heterogeneous with different accessibility issues (for drugs and monitoring), patient populations and clinical practices. Information on adherence is not routinely collected in this cohort which may have an impact on the effectiveness of PI-based regimens. The challenges associated with adhering to PI based regimens are much greater than with NNRTI based regimens, many of which have fixed-dose formulations available. We have limited toxicity data, particularly information on metabolic parameters are not uniformly collected at various time points, thus making it difficult to analyse and conclude the safety of PI based regimens in these settings.
In summary, use of PIs in this cohort is affected by accessibility and affordability issues, particularly for the newer PIs. There is an urgent need to improve access to these drugs as the number of patients needing second line PI-based regimens is expected to increase in the near future. PI based regimens are effective in these settings but increased resources for virological monitoring are also needed to optimise effectiveness through earlier identification of failure.
Acknowledgments
TREAT Asia is a program of The Foundation for AIDS Research, amfAR. The TREAT Asia HIV Observational Database (TAHOD) is supported in part by grants from the U.S. National Institutes of Health’s National Institute of Allergy and Infectious Diseases (NIAID), grant no. U01-AI069907, and the Ministry of Foreign Affairs of the government of The Netherlands. The National Centre in HIV Epidemiology and Clinical Research is funded by The Australian Government Department of Health and Ageing, and is affiliated with the Faculty of Medicine, The University of New South Wales.
Footnotes
This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
References
- 1.Agwu A, Lindsey JC, Ferguson K, Zhang H, Spector S, et al. Analyses of HIV-1 drug-resistance profiles among infected adolescents experiencing delayed antiretroviral treatment switch after initial nonsuppressive highly active antiretroviral therapy. AIDS Patient Care STDS. 2008;22:545–52. doi: 10.1089/apc.2007.0200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Cavalcanti AM, Lacerda HR, Brito AM, Pereira S, Medeiros D, et al. Antiretroviral resistance in individuals presenting therapeutic failure and subtypes of the human immunodeficiency virus type 1 in the Northeast Region of Brazil. Mem Inst Oswaldo Cruz. 2007;102:785–92. doi: 10.1590/s0074-02762007005000109. [DOI] [PubMed] [Google Scholar]
- 3.Crane HM, Van Rompaey SE, Kitahata MM. Initiating highly active antiretroviral therapy with newer protease inhibitors is associated with better survival compared to first-generation protease inhibitors or nevirapine. AIDS Patient Care STDS. 2007;21:920–9. doi: 10.1089/apc.2007.0020. [DOI] [PubMed] [Google Scholar]
- 4.Cozzi-Lepri A, Phillips AN, Ruiz L, Clotet B, Loveday C, et al. Evolution of drug resistance in HIV-infected patients remaining on a virologically failing combination antiretroviral therapy regimen. AIDS. 2007;21:721–32. doi: 10.1097/QAD.0b013e3280141fdf. [DOI] [PubMed] [Google Scholar]
- 5.Duvivier C, Myrto A, Marcelin AG, Ghosn J, Ait-Mohand H, et al. Efficacy and safety of ritonavir/indinavir 100/400 mg twice daily in combination with two nucleoside analogues in antiretroviral treatment-naive HIV-infected individuals. Antivir Ther. 2003;8:603–9. [PubMed] [Google Scholar]
- 6.Gallant JE. Drug resistance after failure of initial antiretroviral therapy in resource-limited countries. Clin Infect Dis. 2007;44:453–5. doi: 10.1086/510752. [DOI] [PubMed] [Google Scholar]
- 7.Gardner EM, Sharma S, Peng G, Hullsiek KH, Burman WJ, et al. Differential adherence to combination antiretroviral therapy is associated with virological failure with resistance. AIDS. 2008;22:75–82. doi: 10.1097/QAD.0b013e3282f366ff. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hicks CB, Cahn P, Cooper DA, Walmsley SL, Katlama C, et al. Durable efficacy of tipranavir-ritonavir in combination with an optimised background regimen of antiretroviral drugs for treatment-experienced HIV-1-infected patients at 48 weeks in the Randomized Evaluation of Strategic Intervention in multi-drug reSistant patients with Tipranavir (RESIST) studies: an analysis of combined data from two randomised open-label trials. Lancet. 2006;368:466–75. doi: 10.1016/S0140-6736(06)69154-X. [DOI] [PubMed] [Google Scholar]
- 9.Marconi VC, Sunpath H, Lu Z, Gordon M, Koranteng-Apeagyei K, et al. Prevalence of HIV-1 drug resistance after failure of a first highly active antiretroviral therapy regimen in KwaZulu Natal, South Africa. Clin Infect Dis. 2008;46:1589–97. doi: 10.1086/587109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Martinez-Cajas JL, Wainberg MA. Antiretroviral therapy: optimal sequencing of therapy to avoid resistance. Drugs. 2008;68:43–72. doi: 10.2165/00003495-200868010-00004. [DOI] [PubMed] [Google Scholar]
- 11.Mootsikapun P, Chetchotisakd P, Anunnatsiri S, Boonyaprawit P. Efficacy and safety of indinavir/ritonavir 400/100 mg twice daily plus two nucleoside analogues in treatment-naive HIV-1-infected patients with CD4+ T-cell counts <200 cells/mm3: 96-week outcomes. Antivir Ther. 2005;10:911–6. [PubMed] [Google Scholar]
- 12.Ortiz R, Dejesus E, Khanlou H, Voronin E, van Lunzen J, et al. Efficacy and safety of once-daily darunavir/ritonavir versus lopinavir/ritonavir in treatment-naive HIV-1-infected patients at week 48. AIDS. 2008;22:1389–97. doi: 10.1097/QAD.0b013e32830285fb. [DOI] [PubMed] [Google Scholar]
- 13.Palella FJ, Jr, Baker RK, Moorman AC, Chmiel JS, Wood KC, et al. Mortality in the highly active antiretroviral therapy era: changing causes of death and disease in the HIV outpatient study. J Acquir Immune Defic Syndr. 2006;43:27–34. doi: 10.1097/01.qai.0000233310.90484.16. [DOI] [PubMed] [Google Scholar]
- 14.Pellegrin I, Wittkop L, Joubert LM, Neau D, Bollens D, et al. Virological response to darunavir/ritonavir-based regimens in antiretroviral-experienced patients (PREDIZISTA study) Antivir Ther. 2008;13:271–9. [PubMed] [Google Scholar]
- 15.Pierone G, Jr, Mieras J, Bulgin-Coleman D, Kantor C, Shearer J, et al. A pilot study of switch to lopinavir/ritonavir (LPV/r) monotherapy from nonnucleoside reverse transcriptase inhibitor-based therapy. HIV Clin Trials. 2006;7:237–45. doi: 10.1310/hct0705-237. [DOI] [PubMed] [Google Scholar]
- 16.Pujades-Rodriguez M, O’Brien D, Humblet P, Calmy A. Second-line antiretroviral therapy in resource-limited settings: the experience of Medecins Sans Frontieres. AIDS. 2008;22:1305–12. doi: 10.1097/QAD.0b013e3282fa75b9. [DOI] [PubMed] [Google Scholar]
- 17.Pulido F, Arribas JR, Delgado R, Cabrero E, González-García J, et al. Lopinavir-ritonavir monotherapy versus lopinavir-ritonavir and two nucleosides for maintenance therapy of HIV. AIDS. 2008;22:F1–9. doi: 10.1097/QAD.0b013e3282f4243b. [DOI] [PubMed] [Google Scholar]
- 18.Ramadhani HO, Thielman NM, Landman KZ, Ndosi EM, Gao F, et al. Predictors of incomplete adherence, virologic failure, and antiviral drug resistance among HIV-infected adults receiving antiretroviral therapy in Tanzania. Clin Infect Dis. 2007;45:1492–8. doi: 10.1086/522991. [DOI] [PubMed] [Google Scholar]
- 19.Schechter M, Nunes EP. Monotherapy with lopinavir/ritonavir. Expert Opin Investig Drugs. 2007;16:735–41. doi: 10.1517/13543784.16.5.735. [DOI] [PubMed] [Google Scholar]
- 20.Sen S, Tripathy SP, Patil AA, Chimanpure VM, Paranjape RS. High prevalence of human immunodeficiency virus type 1 drug resistance mutations in antiretroviral treatment-experienced patients from Pune, India. AIDS Res Hum Retroviruses. 2007;23:1303–8. doi: 10.1089/aid.2007.0090. [DOI] [PubMed] [Google Scholar]
- 21.Sung H, Jung YS, Kang MW, Bae IG, Chang HH, et al. High frequency of drug resistance mutations in human immunodeficiency virus type 1-infected Korean patients treated with HAART. AIDS Res Hum Retroviruses. 2007;23:1223–9. doi: 10.1089/aid.2007.0008. [DOI] [PubMed] [Google Scholar]
- 22.Sungkanuparph S, Manosuthi W, Kiertiburanakul S, Piyavong B, Chumpathat N, et al. Options for a second-line antiretroviral regimen for HIV type 1-infected patients whose initial regimen of a fixed-dose combination of stavudine, lamivudine, and nevirapine fails. Clin Infect Dis. 2007;44:447–52. doi: 10.1086/510745. [DOI] [PubMed] [Google Scholar]
- 23.World Health Organization. Towards Universal Access: Scaling up priority HIv/AIDS interventions in the health sector. WHO/UNAIDS/UNICEF; 2008. p. 2008. [Google Scholar]
- 24.World Bank Organization. Country Profile. [Google Scholar]
- 25.Zhou J, Kumarasamy N. Predicting short-term disease progression among HIV-infected patients in Asia and the Pacific region: preliminary results from the TREAT Asia HIV Observational Database (TAHOD) HIV Med. 2005;6:216–23. doi: 10.1111/j.1468-1293.2005.00292.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Zhou J, Kumarasamy N, Ditangco R, Kamarulzaman A, Lee CK, et al. The TREAT Asia HIV Observational Database: baseline and retrospective data. J Acquir Immune Defic Syndr. 2005;38:174–9. doi: 10.1097/01.qai.0000145351.96815.d5. [DOI] [PMC free article] [PubMed] [Google Scholar]