Abstract
Background
Prior studies using reverse transcriptase inhibitors to treat a human betaretrovirus (HBRV) in patients with primary biliary cholangitis (PBC) resulted in a 21% reduction in alkaline phosphatase (ALP). Herein, we studied the safety and efficacy of combination tenofovir-emtricitabine (TDF/FTC) and lopinavir-ritonavir (LPRr) in PBC patients unresponsive to ursodeoxycholic acid (UDCA).
Methods
A double-blind randomized controlled trial was performed in patients on UDCA for 6 months or more with ALP levels greater than two-fold the upper limit of normal or bilirubin greater than the upper limit of normal. Patients were randomized to daily TDF/FTC 300/200 mg and LPRr 800/200 mg versus identical placebo for 6 months. The primary endpoint was reduction of ALP below 1.67 × ULN or normalization of bilirubin. HBRV DNA levels were assessed in peripheral blood mononuclear cells (PBMC) using digital droplet polymerase chain reaction.
Results
The enrolment was limited to 13 patients because most patients were unable to tolerate LPRr due to the development of gastrointestinal symptoms. No difference in the primary endpoint was achieved. A significant reduction was observed in ALP by 25% (P < 0.05) and in HBRV proviral load (P < 0.05) after 6 months of combination antiretroviral therapy. The majority of patients had diminished levels of LPRr after 6 months’ therapy suggesting inadequate intake of protease inhibitor toward the end of the study.
Conclusions
Combination anti-retroviral therapy resulted in improvement in hepatic biochemistry with reduction in proviral load. The frequency of side effects from LPRr in patients with PBC exceeds the frequency reported for HIV, warranting the search for better tolerated combinations in future studies.
Keywords: combination antiretroviral therapy, human betaretrovirus, primary biliary cholangitis
Introduction
Primary biliary cholangitis (PBC) is a chronic cholestatic disorder characterized by immune destruction of interlobular bile ducts (1–3). The role that autoimmunity plays in the pathogenesis of PBC is unknown and immunosuppressive treatments are not routinely used due to either lack of efficacy or undue side effects (2,3). Ursodeoxycholic acid (UDCA) is the standard of care for PBC patients and the farnesoid X receptor agonist, obeticholic acid (4) has recently been licenced for patients with an inadequate response to UDCA (5,6) or for those intolerant of UDCA (7). Obeticholic acid therapy provides clinically meaningful improvements when bilirubin and alkaline phosphatase (ALP) are used as surrogate markers for disease progression in PBC patients (8).
The etiology of PBC is thought to be related to the effects of environmental triggers in genetically susceptible individuals (9). Our group pursued a viral involvement in PBC after finding serological reactivity to retroviruses (10), electron microscopy evidence of virus-like particles in biliary epithelial cells, and human betaretrovirus (HBRV) sequences in perihepatic lymph nodes and biliary epithelial cells from PBC patients (11–13). Several groups, including our own, reported that the levels of HBRV were below the level of detection in hepatic DNA with a single round of polymerase chain reaction (PCR) (11,14,15), mandating the need to detect viral integrations—considered the gold standard for demonstrating retroviral infection (16). Accordingly, ligation-mediated PCR and next generation sequencing were used to detect over 2,000 HBRV integration sites in patients’ liver, cholangiocyte, and lymph node samples (17). Both HBRV RNA and proviral integrations were also detected in the majority of PBC patients’ lymph nodes and biliary epithelial cells (17).
Prior proof of principle open-label studies and randomized controlled trials (RCT) have been conducted with PBC patients to determine whether antiretroviral therapy (ART) improves clinical outcomes (18). Animal studies have established that the reverse transcriptase (RT) inhibitor zidovudine inhibits the closely related betaretrovirus, mouse mammary tumour virus (MMTV) in vivo (19). A 12-month, open-label pilot study using lamivudine and zidovudine (Combivir) reported histological and biochemical improvement in patients with PBC (20). Then a RCT with lamivudine and zidovudine showed a 21% reduction of ALP from baseline with 6 months therapy but the study failed to achieve a significant difference in the stated endpoints of normalization of ALP, aspartate aminotransferase (AST), or alanine aminotransferase (ALT) (21). However, evidence of virological resistance to therapy was seen in patients experiencing biochemical rebound and viral breakthrough during lamivudine and zidovudine therapy (22).
To find suitable combination ART (cART) against betaretrovirus infection, the HIV protease inhibitor lopinavir boosted with ritonavir (LPRr) was evaluated in vitro and found to inhibit betaretrovirus protease activity (23). Then in vivo studies with NOD.c3c4 mouse model with MMTV cholangitis and autoimmune biliary disease were conducted using different combination RT inhibitors with and without LPRr (24). The regimen with tenofovir (TDF) and emtricitabine (FTC), with LPRr had the optimal impact on lowering ALP and abrogating histological cholangitis coinciding with diminished hepatic MMTV levels (22). In the clinic, a case report documented a PBC patient with HIV and HBRV co-infection who normalized liver tests with TDF/FTC/LPRr within a year (25).
The objectives of this RCT were to study whether reduction in viral load corresponded with clinical improvement, and to investigate the safety, tolerability, and efficacy of TDF/FTC/LPRr in patients with PBC who had an inadequate response to UDCA. While cART resulted in reduction in hepatic biochemistry and viral load, most patients were unable to tolerate LPRr.
Methods
Study design
A multi-centre, double-blind, cross-over, placebo-controlled trial was planned to evaluate the response of patients with PBC to cART. PBC patients on UDCA meeting enrolment criteria were randomized 1:1 to the treatment arm of TDF/FTC (300/200 mg daily) and LPRr (400/100 mg twice daily) or an identical placebo. A dynamic computerized schedule was used by the GI and Liver Disease Research study group to ensure equivalent randomization of treatment and placebo. Patients were assessed at month 1, month 3 and at the end of the 6 month study. The primary endpoint was reduction of ALP to less than 1.67 × ULN or normalization of bilirubin (6,8).
The study protocol (ClinicalTrials.gov NCT01614405) and subsequent amendments were reviewed and approved by Health Canada and the Human Ethics Review Board at the University of Alberta. All patients read and signed the consent form and were recruited and followed up at the University of Alberta site. The study was conducted according to good clinical practice guidelines in keeping with the Declaration of Helsinki. Trial monitoring and quality assurance were conducted by the Quality Management in Clinical Research Department at the University of Alberta.
Patient population
Patients with PBC older than 18 years of age and non-responders to UDCA were recruited for the study. The diagnosis of PBC was confirmed based on the combination of at least two of the three following criteria: biochemical evidence of cholestasis with ALP elevation; serological presence of AMA; and histopathological evidence of non-suppurative cholangitis. All patients were maintained on UDCA at a dose of 13–15 mg/kg for at least 6 months before enrolment and during the whole period of the study. The intention was to enrol approximately 80 patients based on the sample size calculations (see Appendix 1).
The inclusion criterion for our clinical trial was lack of response to UDCA after at least 6 months of therapy defined as elevated ALP greater than 2 × ULN or abnormal bilirubin. The major exclusion criteria included elevated plasma AST or ALT greater than 5 × ULN, creatinine clearance less than 70 mL/min, the use of immunosuppression or other study drug within 6 months prior to screening, the presence of significant portal hypertension, Child–Pugh class B or C cirrhosis, or other liver diseases (see Appendix 1).
Efficacy analyses
The primary endpoint was either reduction of ALP to 1.67 × ULN or normalization of bilirubin at 6 months of intervention. (26) The secondary and exploratory endpoints included the change in serum ALP, ALT, and total bilirubin over the course of the study. Changes in symptoms were recorded using the PBC-40 score to assess six symptom domains relating to fatigue, itch, cognitive symptoms, social and emotional symptoms, and other symptoms. (27,28) The higher PBC-40 domain scores indicate a diminished quality of life and scores range as follows: general symptoms (range 7–35), itch (range 3–15), fatigue (range 11–55), cognitive function (range 6–30), social (range 10–50), and emotional (range 3–15). Safety assessments included physical examinations and monitoring adverse events (AE) and vital signs, as well as laboratory testing with CBC, lipid profile, and creatinine to assess renal impairment from TDF. Viral load was assessed in peripheral blood mononuclear cell DNA levels using digital droplet (dd) PCR to measure HBRV proviral DNA in peripheral blood mononuclear cells (PBMC).
Digital droplet PCR (ddPCR)
PBMC DNA was isolated using the QiaAmp DNA blood mini kit (Qiagen). Each 20 µl ddPCR reaction setup consisted for 10 µL of 2 ddPCR Supermix for probes without dUTP (Bio-Rad, Mississauga, ON), 200 ng of patient PBMC DNA and RNase/DNase free H2O, 900 nM pol forward primer 5′ CAC ATG ACC TCT ACC ACA CCA C 3′ and pol reverse primer 5′ TGA CTG CCA CAA TTT CAG CCT 3′ and 450 nM probe targeting the pol gene 5′ HEX ACG GGT CAG /ZEN/ CAA ATG GCC GTT CG /3IABkFQ/ including the internal ZEN quencher and the 3’ Iowa Black FQ quencher (International DNA Technologies, Toronto, Ontario). The reaction setup was performed using 20 µL reaction mixture and 70 µL of droplet generation oil loaded into the droplet generator. The droplets were transferred to a 96 well Eppendorf PCR plate and amplified in an Applied Biosystems 2720 thermal cycler. The cycling conditions were 95°C for 10 minutes, 40 cycles of denaturation 94°C for 10 minutes, and annealing/extension at 58°C for 1 minute, followed by a final enzyme deactivation step at 98°C for 10 minutes. The positive droplets were analyzed on the QX200 droplet reader, the data was processed using the QuantaSoft Software (Bio-Rad Laboratories, Inc.) and the values expressed by calculating the proviral genome equivalents per µg of PBMC DNA.
Serial PBC patient PBMC were collected for DNA extraction and each DNA sample was evaluated on three separate occasions. The median values from the three replicates were used for the final analyses. Negative control PBMC samples were derived from liver disease patients without PBC (n = 36) and healthy patients (n = 15) to establish the lower limit of detection above background signal. Each negative control sample was tested a median of three times (range 1–6) and samples confirmed negative for HBRV proviral DNA were used as to determine a baseline for distinguishing signal from background. Serial dilutions of proviral betaretrovirus plasmid DNA were added to 200 ng of genomic DNA in a range from 10 (4) copies down to one viral genome copy and these were used to calibrate the assay. The lower limit of detection was established as 15 proviral genome equivalents/µg PBMC DNA (equal to three HBRV genomes per ddPCR assay) that represented a positive viral load which consistently detected the presence and quantified the HBRV proviral DNA above the background signal generated by negative controls. However, negative readings below the 15 proviral genome equivalents/µg PBMC DNA level may still contain betaretrovirus DNA but fall below the limits of detection of this ddPCR assay.
Statistical analysis
Initial analyses were performed using Shapiro–Wilk test to assess whether biochemical values and PBC-40 scores were normally distributed for both treatment arms (29,30). As this was found to be true for some variables only, non-parametric statistical methods were applied by comparing parameters before and after 6 months of study using Wilcoxon signed-rank test for paired data. Comparisons of the parameters between treatment and placebo groups at 6 months and other time points in the OLE were performed using the Mann–Whitney U test. Proportions were compared using the Fisher exact test. Data were presented as median (range) and number (%) and analyses were performed using GraphPad Prism version 6.0 (GraphPad Software, San Diego, CA) and IBM SPSS, version 19.0 (IBM Corp., Armonk, NY) with double sided P-value ≤ 0.05 to indicate statistical significance.
Results
Study patients
The enrolment was limited to one centre where only 13 patients were randomized. Protracted gastrointestinal symptoms were experienced by most patients who were subsequently found to be in the treatment arm on LPRr. This observation resulted in cessation of enrolment and a decision not to continue the RCT. The study population consisted of eight PBC patients in the placebo arm and five patients on cART. There were no significant clinical or biochemical differences observed in the PBC patients between treatment and placebo arms at baseline (Table A.1). Two patients in the placebo arm discontinued the RCT; one patient developed a serious adverse event and the second was an incorrect enrolment with Child–Pugh B, who went for a liver transplant work-up (Figure 1). As a result, paired parameters were available for five patients in the treatment arm and six patients in the placebo arm for the double-blind portion of the RCT (Table 1).
Figure 1:
Patient disposition flow chart (CONSORT) for double-blind trial
Table 1:
Biochemical response to antiretroviral therapy
| Baseline | Month 3 | Month 6 | Change (baseline–month 6) | Change (baseline–month 6), P-value | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| TDF/FTC/LPRr | Placebo | TDF/FTC/LPRr | Placebo | TDF/FTC/LPRr | Placebo | TDF/FTC/LPRr | Placebo | TDF/FTC/ LPRr | Placebo | |
| Parameter | (n = 5) | (n = 6) | (n = 5) | (n = 6) | (n = 5) | (n = 6) | (n = 5) | (n = 6) | (n = 5) | (n = 6) |
| ALP U/L | 546 (290–787) | 355 (245–461) | 484 (307–646) | 369 (241–534) | 410 (275–660) | 315 (231–506) | –136 | –40 | ||
| 0.043 | 0.249 | |||||||||
| ALP x ULN | 4.20 (2.23–6.05) | 2.73 (1.88–3.55) | 3.72 (2.36–4.97) | 2.83 (1.85–4.11) | 3.15 (2.12–5.08) | 2.42 (1.78–3.89) | –1.05 | –0.31 | ||
| Bilirubin µmol/L | 13 (8–37) | 29 (9–51) | 28 (8–39) | 23 (9–28) | 21 (16–28) | 17 (10–40) | +8 | –12 | ||
| 0.892 | 0.345 | |||||||||
| Bil x ULN | 0.65 (0.40–1.85) | 1.45 (0.45–2.55) | 1.40 (0.4–1.95) | 1.14 (0.45–1.40) | 1.05 (0.80–1.40) | 0.85 (0.50–2) | +0.4 | –0.6 | ||
| ALT U/L | 103 (62–165) | 115 (51–141) | 113 (88–162) | 126 (45–150) | 89 (61–216) | 101 (49–176) | –14 | –14 | 0.893 | 0.917 |
| AST U/L | 101 (48–127) | 89 (39–210) | 89 (69–133) | 89 (38–228) | 73 (69–190) | 108 (36–159) | –28 | +19 | 0.109 | 0.917 |
| GGT U/L | 412 (190–682) | 513 (230–1236) | 460 (336–583) | 442 (236–1,407) | 502 (279–593) | 374 (124–1,508) | +90 | –139 | 0.197 | 0.249 |
Note: Values expressed as median (min–max). Analyses are based on Mann–Whitney U test.
Abbreviations: TDF/FTC/LPRr = combination tenofovir-emtricitabine (TDF/FTC) and lopinavir-ritonavir (LPRr); ALP = alkaline phosphatase; ULN = upper limit of normal range; ALT = alanine aminotransferase; AST = aspartate aminotransferase; GGT = gamma-glutamyl transferase
HBRV quantification
The lower limit of detection of HBRV proviral was calculated to be 15 proviral genome equivalents/µg PBMC DNA, above which no spurious signal was observed in the negative controls. Using this cut-off level, 2 of the 36 non-PBC liver disease controls (6%) and none of the 15 healthy control patients tested positive by ddPCR for HBRV. At baseline, 7 of 11 (64%) patients in the RCT tested HBRV-positive and subsequently, 2 of the 4 baseline-negative patients developed a positive HBRV level during the study to provide a total 81% of the study population that were recorded as HBRV positive. One of the HBRV-negative patients remained undetectable by ddPCR and the other patient lacked samples for follow-up testing.
Endpoints and efficacy assessments
None of the patients in the treatment arm reached the primary biochemical endpoint at 6 months as compared with one patient in the placebo arm, who experienced a slight reduction in ALP (5.7%). Initially, the patients on cART experienced worsening of their hepatic biochemistry after one month’s intervention with a median increase of 41 U/L in ALP levels and 21 U/L in ALT (Figure 2) as well as a doubling of bilirubin by 3 months (Table 1). Three of the five patients experienced a rise in ALP and all five increased their ALT after one month of treatment (Figure 3). After 6 months’ treatment with TDF/FTC/LPRr, the patients experienced a significant reduction in median ALP from baseline (Figure 2; median change in ALP −80 U/L [range −215 to −15], P < 0.05), but no significant changes in ALT or bilirubin. With regard to symptomatic assessment, no significant changes were observed in any of the six domains of the PBC-40 from baseline to 6 months following intervention (Table 2).
Figure 2:
Change from baseline in ALP, ALT, and proviral load comparing TDF/FTC/LPRr vs. placebo patients completing the double-blind randomized controlled trial
A, B: patients initially experienced an increase in ALP and ALT levels from baseline after 1 month of TDF/FTC/LPRr therapy and then experienced a median 80 U/L reduction in ALP (P < .05). C: A significant reduction in proviral load in PBMC DNA was observed in patients treated with TDF/FTC/LPRr for 6 months (P < .05).
Note: Data is expressed as median (and range) and analysis was performed using the Mann–Whitney U test and Wilcoxon signed-rank test
ALP = alkaline phosphatase; ALT = alanine aminotransferase; TDF/FTC/LPRr = combination tenofovir-emtricitabine (TDF/FTC) and lopinavir-ritonavir (LPRr); PBMC = peripheral blood mononuclear cells
Figure 3:
Individual biochemical and virological responses in double-blind trial comparing TDF/FTC/LPRr (n = 5) vs. placebo (n = 6)
Changes in (A) ALP, (B) ALT, and (C) proviral load after 6 months of intervention
* Patient 010 with LPR levels < 0.01 mg/L after 6 months of therapy (see Table A.2)
TDF/FTC/LPRr = combination tenofovir-emtricitabine (TDF/FTC) and lopinavir-ritonavir (LPRr); ALP = alkaline phosphatase; ALT = alanine aminotransferase
Table 2:
PBC-40 symptomatic response, TDF/FTC/LPRr versus placebo
| TDF/FTC/LPRr (n = 5) | Placebo (n = 6) | |||||
|---|---|---|---|---|---|---|
| PBC-40 domain | Baseline | Month 6 | Changes over time (baseline–month 6) P-value | Baseline | Month 6 | Changes over time (baseline–month 6) P-value |
| Symptoms | 18 (13–23) | 17 (15–28) | 0.109 | 19.5 (15–28) | 22 (11–23) | 0.500 |
| Itch | 6.5 (6–8) | 9.5 (4–45) | 0.285 | 8.5 (5–13) | 7 (6–14) | 0.916 |
| Fatigue | 33 (22–41) | 37.5 (27–40) | 0.285 | 31.5 (16–44) | 33 (14–46) | 0.833 |
| Cognitive | 15 (10–18) | 18.5 (12–22) | 0.180 | 14.5 (6–19) | 13 (9–22) | 0.686 |
| Emotional | 9 (6–10) | 9 (7–11) | 0.157 | 6.5 (3–12) | 9 (4–12) | 0.357 |
| Social | 26 (17–31) | 31 (19–34) | 0.109 | 23.5 (16–45) | 25.5 (14–45) | 0.854 |
| Total | 111.5 (75–122) | 122.5 (91–143) | 0.109 | 104.5 (67–149) | 112.5 (58–160) | 0.917 |
Note: Values are expressed as median (min–max). Analyses are based on Wilcoxon signed-rank test.
In response to the antiviral therapy, patients receiving TDF/FTC/LPRr experienced a significant fall in HBRV levels after 6 months with a median reduction in proviral load from 120 (range 14–290) at baseline to 32 (range 7–59) HBRV proviral genome equivalents/µg PBMC DNA (P < 0.05). In contrast, no significant changes were observed in the placebo arm (Figure 2C). Notably, one patient demonstrated a ~33% reduction in all hepatic biochemistry studies coinciding with a marked reduction in HBRV levels (Figure 3). Yet in the placebo arm, HBRV levels were found to vary over the 6-month period of the RCT without any drug intervention. For example, one patient was found to have undetectable viral load at baseline and then 290 HBRV proviral genomes per 1 µg PBMC DNA 6 months later without obvious changes in her hepatic biochemistry (Figure 3).
Side effects and compliance
A serious adverse event was recorded in a patient from the placebo arm who dropped out of the study at 4 months due to variceal bleeding. The reported side effects of LPRr were experienced in 80% of patients in the treatment arm who developed gastrointestinal intolerance related to a combination of symptoms that included nausea, vomiting, abdominal pain, bloating, weight loss, constipation, and diarrhea (Table 3). Notably, 50% of the placebo patients also reported gastrointestinal intolerance but of shorter duration.
Table 3:
Frequency of serious adverse events and adverse events occurring during the study
| No. (%) of patients | ||
|---|---|---|
| Adverse event | TDF/FTC/LPRr (n = 5) | Placebo (n = 6) |
| Gastrointestinal intolerance (combinations of the following: loss of appetite, nausea, vomiting, epigastric pain or reflux, abdominal cramps, flatulence, dysgeusia, diarrhea, change in bowel pattern) | 4 (80) | 3 (50) |
| Fatigue | 1 (20) | 1 (17) |
| Intermittent shortness of breath | 1 (20) | 1 (17) |
| Flu-like symptoms | 1 (20) | – |
| Increased frequency of headaches | – | 1 (17) |
| Generalized arthralgias | – | 2 (33) |
| Dehydration | – | 2 (33) |
| Hypercholesterinemia | – | 1 (17) |
| Strep throat | – | 1 (17) |
| Anemia | – | 1 (17) |
| Hot flushes | – | 1 (17) |
| Leg cramps | – | 1 (17) |
| Urinary tract infection | – | 1 (17) |
| Dry mouth | – | 1 (17) |
| Change in sleep pattern | – | 1 (17) |
| Serious adverse events (hematemesis from esophageal varices) | – | 1 (17) |
Abbreviations: TDF/FTC/LPRr = combination tenofovir-emtricitabine (TDF/FTC) and lopinavir-ritonavir (LPRr)
Serum was assessed for lopinavir and ritonavir levels to determine whether variations in compliance or drug absorption impacted on biochemical outcomes (Table A.2). All but one patient had considerably reduced serum levels of both lopinavir and ritonavir levels at 6 months as compared with 3 months and the remaining patient had diminished levels at 3 months. One of the patients in the RCT who reported vomiting tablets, was found to have undetectable levels of lopinavir and ritonavir after 6 months of therapy coinciding in biochemical rebound of both ALP and ALT (Figure 3).
Conclusions
While we observed that reduction of HBRV levels in PBMC DNA was associated with biochemical improvement in PBC patients treated with combination ART, this study was beleaguered by the lack of tolerability of the study drug. Accordingly, we did not enrol a sufficient sample size, many patients failed to achieve adequate serum lopinavir levels and primary endpoints were not met. The known hepatotoxicity of LPRr was manifest early with drug-induced liver injury after 1 month’s treatment and the impact of patient well-being, as recorded by the PBC-40 symptoms’ scores. It is unknown why LPRr was so poorly tolerated with the majority of PBC patients developing gastrointestinal intolerance. In comparison with other populations, the discontinuation rates in HIV trials range from 7%–10% in treatment of HIV infection and approximately 20% for post-exposure prophylaxis (31–33).
Notably, one patient in Figure 3 developed a biochemical response associated with a reduction in viral load, without experiencing undue drug toxicity (red). Most patients developed worsening liver tests within the first month before the group collectively experienced a 25% reduction in ALP. The kinetics of improvement in ALP is very different than that observed with obeticholic acid, where patients experienced a rapid fall within the first month and then the ALP levelled out thereafter in months 3–6 (6). Indeed, the slow fall in ALP with TDF/FTC/LPRr may be related to the incremental reduction in HBRV levels and the ensuing open-label observational study supports hypothesis (34). Indeed, it is apparent from the observed biochemical and virological trends and from other reports that a longer period of therapy may have been beneficial for those who could tolerate LPRr (18,25,34). Nevertheless, these data represent an incremental improvement as PBC patients achieved a 20% reduction of ALP in a prior 6-month RCT with lamivudine and zidovudine (21). For patients with HIV, the introduction of protease inhibitors resulted in a marked improvement in treatment because RT inhibitor therapy was associated with the development of resistance associated variants within weeks of starting therapy (35). Indeed, we observed such resistance associated variants in the NOD.c3c4 mouse (22) and with lamivudine and zidovudine therapy in PBC patients (18), which provided impetus to test more robust combination ART regimens against HBRV.
Another important aspect of this study was the ability to monitor viral load in response to ART, which was lacking in our prior studies. The ddPCR methodology can be slightly more sensitive than quantitative PCR but the main advantage is consistency, precision and reproducibility in measuring low copy number (36–38). For example, ddPCR has been shown to provide a 20-fold improvement in accuracy for quantifying residual HIV DNA verses quantitative PCR (37). Notably, ddPCR delivers comparable quantitative data for measuring low copy number of HIV RNA or residual proviral HIV DNA in patients on cART (37). Accordingly, we chose to measure HBRV DNA for assessing response to cART as the copy number of this intracellular virus is low and proviral DNA is more stable than HBRV RNA. (However, we have not yet compared the measurement of HBRV RNA versus DNA levels). To avoid false positive results, a conservative cut-off value for detection of HBRV DNA was instituted using multiple HBRV negative samples to avoid background false positive signal. With this cut-off, 81% of the trial patients had evidence of HBRV on serial testing and proviral load was observed to vary without any intervention. Furthermore, a significant reduction in HBRV DNA levels was observed in patients with TDF/FTC/LPRr and the decrease in proviral load coincided with improvement in hepatic biochemistry. However, it may become relevant to measure HBRV RNA to more closely correlate viral load with biochemistry in future studies.
In summary, we report as a proof of principle that a small group of individuals on cART experienced reduction in hepatic biochemistry and HBRV proviral load. LPRr is not an option for future PBC clinical trials and as gastrointestinal intolerance is a class effect of HIV protease inhibitors, it would be judicious to evaluate better tolerated HIV integrase inhibitors instead. These agents have largely replaced the protease inhibitors as front-line therapy and some HIV integrase inhibitors, such as raltegravir and dolutegravir, have broad spectrum antiretroviral against other retroviruses. Such studies are warranted as HBRV suppression appears to correlate with clinical improvement in patients with PBC.
Appendix 1: Supplemental Methods
Oversight
Study conception and trial design, was conducted in collaboration with the academic authors. LS provided in house management and advice concerning antiretroviral use. ALM had the final responsibility to submit the manuscript after all authors reviewed and made amendments.
The Data and Safety Management Board members were Kurt Williams, MD and Stuart Skinner, MD, and the onsite medical monitoring was provided by the Quality Management in Clinical Research Department at the University of Alberta. Patients were recruited and managed by the Gastrointestinal and Liver Disease Research (GILDR) group and data collection was performed by EpiCore at the University of Alberta.
Bellwyck Packaging Ltd., Mississauga, Ontario, received medications and identical placebo from Abbvie and Gilead Sciences. They coded and packaged the identical bottles for distribution.
Inclusion and exclusion criteria
Major inclusion criteria
Patients 18 years old of either sex will be recruited for this study.
Elevated ALP after 6 months UDCA therapy × upper limit of normal or abnormal bilirubin.
Positive serum AMA or Liver biopsy histology compatible with PBC.
Maintained on UDCA at a dose of 13–15 mg/kg for 6 or more months or intolerant to UDCA.
Patients must read and sign informed consent form.
Major exclusion criteria
Patients with baseline AST or ALT > 5 × ULN.
Patients who have altered dose of any medications used to treat PBC (such as UDCA) or the use of colchicine, corticosteroids, azathioprine, chlorambucil, methotrexate, or D-penicillamine within the last 6 months.
Advanced liver disease or esophageal varices, INR > 1.2 (upper limit of normal), Albumin < 35 g/L (lower limit of normal), platelets < 120,000/mm3, Child–Pugh class B or C cirrhosis, presence of varices or previous variceal hemorrhage, spontaneous encephalopathy, ascites, or need for liver transplantation [as sicker patients are more at risk from complications of therapy].
Patients with a secondary diagnosis such as HIV, viral hepatitis, drug-induced liver injury, extrahepatic biliary obstruction, primary sclerosing cholangitis, metabolic liver diseases or alcoholic liver disease, or regular use of more than 30 g of alcohol per day in the last year. Clinically apparent pancreatitis or with a predicted survival of less than 3 years from malignant or other potentially life-threatening disease.
An ultrasound showing lack of hepatic mass within the last year in patients with cirrhosis.
Previous allergic reaction to study medications.
-
Creatinine clearance less than < 70 mL/min using the Cockcroft–Gault equation:
Pregnancy or breastfeeding a child.
Young sexually active patients not using contraception.
Safety
Patients on TDF/FTC may develop renal insufficiency and a dose reduction with TFV is recommended in renal impairment with a creatinine clearance 30–49 mL/min. Creatinine was serially monitored accordingly. LPVr is reportedly well tolerated in patients with liver disease but up to 30% of patients may develop gastrointestinal intolerance, occasional worsening of diabetes, hyperlipidemia, and rare allergic responses (39). All protease inhibitors are associated with a ~5% risk of grade 3 or 4 hepatotoxicity; accordingly, AST and ALT levels were measured 3-monthly, and fasting lipids and glucose were assessed between 3 and 6 months after starting antiviral treatment and then annually thereafter depending on risk factors for heart disease.
Sample size
Our prior studies suggested that 3.5% of PBC patients may spontaneously drop the ALP below 1.67× ULN from more than 2-fold ULN [however, larger clinical trials evaluating obeticholic acid have subsequently reported that 9%–10% of PBC patients may spontaneously achieve the endpoint without intervention (5, 6)]. We assumed that more than 25% of patients may develop a reduction in ALP < 1.67× ULN based on our own and other’s PBC patient observations (25) and our animal data (22). Therefore, calculations based on the assumption of active group response rates (> 25%) and spontaneous response rates (< 3.5%) suggested a total sample size of 68–72 patients for a power of 0.80 with a two-sided P-value of 0.05 using a test for two independent proportions. To account for a 10% dropout from the study and to generate additional virological and histological data, we intended to enrol 80 patients.
Lopinavir levels
Lopinavir and ritonavir Cmin concentration was measured 12 hours post-dose (with BID dosing) by HPLC using UV detection in serum collected and stored at −20°C (service kindly performed by Dr. N. Sheehan, Faculty of Pharmacy, University of Montreal, Montreal, Quebec) as previously described (40). In the treatment of HIV, the optimal lopinavir targets can vary depending on the patient. For those with HIV infection without past virological failure to protease inhibitors and with viruses without protease mutations that confer resistance to lopinavir, a minimum concentration greater than 1 mg/L is essential to prevent resistance associated variants, whereas greater than 2 mg is preferable (41). There are no data to guide treatment of HBRV infection with lopinavir-ritonavir, however.
Table A.1:
Baseline characteristics of patients in treatment and placebo arm
| Parameters | TDF/FTC/LPRr (n = 5) |
Placebo (n = 8) |
P-value |
|---|---|---|---|
| Age, y | 53 (43–59) | 50 (39–57) | .215 |
| Female | 4 (80.0) | 7 (87.5) | 1 |
| Years since diagnosis | 7 (4–12) | 3 (2–12) | .165 |
| AMA positive | 4 (80.0) | 8 (100) | .385 |
| ALP, U/L 0.012738 | 546 (290–787) | 393 (245–510) | .124 |
| ALP, × ULN | 4.2 (2.2–6.1) | 3.0 (1.9–3.9) | .124 |
| ALP, > 1.67 × ULN | 5 (100) | 8 (100) | 1 |
| Bilirubin, µmol/L | 13 (8–37) | 29 (9–95) | .610 |
| Bilirubin, × ULN | 0.7 (0.4–1.9) | 1.5 (0.5–4.8) | .610 |
| ALT, × /L | 103 (62–165) | 115 (51–165) | 1 |
| ALT, × ULN | 2.1 (1.2–3.3) | 2.3 (1.0–3.3) | 1 |
| AST, U/L | 101 (48–127) | 92 (39–210) | .944 |
| AST, × ULN | 2.5 (1.2–3.2) | 2.3 (1.0–5.3) | .944 |
| Albumin, g/L | 40 (39–45) | 38 (33–43) | .124 |
| Albumin, × LLN | 1.1 (1.1–1.3) | 1.1 (0.9–1.2) | .124 |
| Platelet count, 109/L | 249 (195–358) | 171 (145–318) | .093 |
| Platelet count, × LLN | 1.8 (1.4–2.6) | 1.2 (1.0–2.3) | .093 |
| Creatinine, µmol/L | 68 (56–80) | 62 (40–86) | .509 |
| IgM, g/L | 3.6 (2.6–4.4) | 2.8 (1.3–5.3) | .826 |
| IgM, × ULN | 1.2 (0.9–1.5) | 0.9 (0.4–1.8) | .826 |
| INR | 1.0 (0.9–1.0) | 1.0 (0.8–1.1) | .826 |
| Cholesterol, µmol/L | 8.0 (5.0–10.5) | 6.6 (5.0–8.1) | .254 |
Note: Values are expressed as median (min–max) or number (%). Analyses are based on Mann–Whitney U test (2-tailed) and Fisher exact test.
Abbreviations: TDF/FTC/LPRr = combination tenofovir-emtricitabine (TDF/FTC) and lopinavir-ritonavir (LPRr); AMA = anti-mitochondrial antibodies; ALP = alkaline phosphatase; ULN = upper limit of normal range; ALT = alanine aminotransferase; AST = aspartate aminotransferase; LLN = lower limit of normal range; INR = international normalized ratio
Table A.2:
Serum levels of lopinavir and ritonavir in patients on active therapy (mg/L)
| Study ID | Compound | Month 3 | Month 6 |
|---|---|---|---|
| 001 | Lopinavir | 17.95 | 3.19 |
| Ritonavir | 1.97 | 0.06 | |
| 003 | Lopinavir | 12.65 | 2.97 |
| Ritonavir | 1.45 | 0.13 | |
| 010 | Lopinavir* | 21.50 | < .01 |
| Ritonavir | 1.32 | < .01 | |
| 012 | Lopinavir | 3.27 | 6.44 |
| Ritonavir | 0.16 | 0.22 | |
| 017 | Lopinavir | 18.10 | 3.78 |
| Ritonavir | 1.68 | 0.11 |
* Levels < 1 mg/L considered subtherapeutic
Funding Statement
The study was supported by the Canadian Institutes for Health research (MOP 114998) and the Canadian Liver Foundation.
Funding:
The study was supported by the Canadian Institutes for Health research (MOP 114998) and the Canadian Liver Foundation. Gilead and Abbvie provided trial medications and placebo free of charge. The Quality Management in Clinical Research Department at the University of Alberta performed the monitoring free of charge.
Disclosures:
Andrew Mason is on the advisory board and has grant funding from Intercept Pharma; he has received medications and placebo for this trial from Gilead and Abbvie.
References
- 1.Cheung AC, Montano-Loza A, Swain M, et al. Time to make the change from ‘primary biliary cirrhosis’ to ‘primary biliary cholangitis’. Can J Gastroenterol Hepatol. 2015;29(6):293. 10.1155/2015/764684. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.European Association for the Study of the Liver; Hirschfield GM, Beuers U, Corpechot C, et al. EASL Clinical Practice Guidelines: The diagnosis and management of patients with primary biliary cholangitis. J Hepatol. 2017; 67(1):145–72. 10.1016/j.jhep.2017.03.022. Medline: [DOI] [PubMed] [Google Scholar]
- 3.Lindor KD, Gershwin ME, Poupon R, et al.; American Association for Study of Liver Diseases. Primary biliary cirrhosis. Hepatology. 2009;50(1):291–308. 10.1002/hep.22906. Medline: [DOI] [PubMed] [Google Scholar]
- 4.Trauner M, Fuchs CD, Halilbasic E, Paumgartner G. New therapeutic concepts in bile acid transport and signaling for management of cholestasis. Hepatology. 2017;65(4):1393–1404. 10.1002/hep.28991. Medline: [DOI] [PubMed] [Google Scholar]
- 5.Hirschfield G, Mason A, Luketic V, et al. The farnesoid X receptor agonist obeticholic acid is efficacious in patients with primary biliary cirrhosis failing therapy with ursodeoxycholic acid: results of a multi-center, randomized, placebo-controlled trial. N Engl J Med. 2013. [Google Scholar]
- 6.Nevens F, Andreone P, Mazzella G, et al. A placebo-controlled trial of obeticholic acid in primary biliary cholangitis. N Engl J Med. 2016;375(7):631–43. 10.1056/NEJMoa1509840. Medline: [DOI] [PubMed] [Google Scholar]
- 7.Kowdley KV, Luketic V, Chapman R, et al. A randomized trial of obeticholic acid monotherapy in patients with primary biliary cholangitis. Hepatology. 2017;67(5):1890–1902. 10.1002/hep.29569. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lammers WJ, van Buuren HR, Hirschfield GM, et al. Levels of alkaline phosphatase and bilirubin are surrogate end points of outcomes of patients with primary biliary cirrhosis: an international follow-up study. Gastroenterology. 2014;147(6):1338–49 e5. 10.1053/j.gastro.2014.08.029. Medline: [DOI] [PubMed] [Google Scholar]
- 9.Hirschfield GM, Chapman RW, Karlsen TH, Lammert F, Lazaridis KN, Mason AL. The genetics of complex cholestatic disorders. Gastroenterology. 2013;144(7):1357–74. 10.1053/j.gastro.2013.03.053. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Mason A, Xu L, Guo L, et al. Detection of retroviral antibodies in primary biliary cirrhosis and other idiopathic biliary disorders. Lancet. 1998;351(9116):1620–4. 10.1016/S0140-6736(97)10290-2. Medline: [DOI] [PubMed] [Google Scholar]
- 11.Sharon D, Mason A. Role of novel retroviruses in chronic liver disease: assessing the link of human betaretrovirus with primary biliary cirrhosis. Curr Infect Dis Rep. 2015;17(2): Article 4. 10.1007/s11908-014-0460-7. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Xu L, Shen Z, Guo L, et al. Does a betaretrovirus infection trigger primary biliary cirrhosis? Proc Natl Acad Sci U S A. 2003;100(14):8454–9. 10.1073/pnas.1433063100. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Xu L, Sakalian M, Shen Z, Loss G, Neuberger J, Mason A. Cloning the human betaretrovirus proviral genome from patients with primary biliary cirrhosis. Hepatology. 2004;39(1):151–6. 10.1002/hep.20024. Medline: [DOI] [PubMed] [Google Scholar]
- 14.Selmi C, Ross SR, Ansari AA, et al. Lack of immunological or molecular evidence for a role of mouse mammary tumor retrovirus in primary biliary cirrhosis. Gastroenterology. 2004;127(2):493–501. 10.1053/j.gastro.2004.05.033. Medline: [DOI] [PubMed] [Google Scholar]
- 15.Johal H, Scott GM, Jones R, Camaris C, Riordan S, Rawlinson WD. Mouse mammary tumour virus-like virus (MMTV-LV) is present within the liver in a wide range of hepatic disorders and unrelated to nuclear p53 expression or hepatocarcinogenesis. J Hepatol. 2009;50(3):548–54. 10.1016/j.jhep.2008.10.026. Medline: [DOI] [PubMed] [Google Scholar]
- 16.Voisset C, Weiss RA, Griffiths DJ. Human RNA “Rumor” viruses: the search for novel human retroviruses in chronic disease. Microbiol Mol Biol Rev. 2008;72(1):157–96. 10.1128/MMBR.00033-07. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wang W, Indik S, Wasilenko ST, et al. Frequent proviral integration of the human betaretrovirus in biliary epithelium of patients with autoimmune and idiopathic liver disease. Aliment Pharmacol Ther. 2015;41(4):393–405. 10.1111/apt.13054. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Lytvyak E, Montano-Loza AJ, Mason AL. Combination antiretroviral studies for patients with primary biliary cirrhosis. World J Gastroenterol. 2016;22(1):349ߝ60. 10.3748/wjg.v22.i1.349. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Held W, Waanders GA, Acha-Orbea H, MacDonald HR. Reverse transcriptase-dependent and-independent phases of infection with mouse mammary tumor virus: implications for superantigen function. J Exp Med. 1994;180(6):2347–51. 10.1084/jem.180.6.2347. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Mason AL, Farr GH, Xu L, Hubscher SG, Neuberger JM. Pilot studies of single and combination antiretroviral therapy in patients with primary biliary cirrhosis. Am J Gastroenterol. 2004;99(12):2348–55. 10.1111/j.1572-0241.2004.40741.x. Medline: [DOI] [PubMed] [Google Scholar]
- 21.Mason AL, Lindor KD, Bacon BR, Vincent C, Neuberger JM, Wasilenko ST. Clinical trial: randomized controlled study of zidovudine and lamivudine for patients with primary biliary cirrhosis stabilized on ursodiol. Aliment Pharmacol Ther. 2008;28(7):886–94. 10.1111/j.1365-2036.2008.03799.x. Medline: [DOI] [PubMed] [Google Scholar]
- 22.Sharon D, Chen M, Zhang G, et al. Impact of combination antiretroviral therapy in the NOD.c3c4 mouse model of autoimmune biliary disease. Liver Int. 2015;35(4):1442–50. 10.1111/liv.12699. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Montano-Loza AJ, Wasilenko S, Bintner J, Mason AL. Cyclosporine A inhibits in vitro replication of betaretrovirus associated with primary biliary cirrhosis. Liver Int. 2010;30(6):871–77. 10.1111/j.1478-3231.2010.02257.x. Medline: [DOI] [PubMed] [Google Scholar]
- 24.Zhang G, Chen M, Graham D, et al. Mouse mammary tumor virus in anti-mitochondrial antibody producing mouse models. J Hepatol. 2011;55(4):876–84. 10.1016/j.jhep.2011.01.037. Medline: [DOI] [PubMed] [Google Scholar]
- 25.Schembri G, Schober P. Killing two birds with one stone. Lancet. 2011;377(9759):96. 10.1016/S0140-6736(10)61343-8. Medline: [DOI] [PubMed] [Google Scholar]
- 26.Silveira MG, Brunt EM, Heathcote J, Gores GJ, Lindor KD, Mayo MJ. American Association for the Study of Liver Diseases endpoints conference: design and endpoints for clinical trials in primary biliary cirrhosis. Hepatology. 2010;52(1):349–59. 10.1002/hep.23637. Medline: [DOI] [PubMed] [Google Scholar]
- 27.Jacoby A, Rannard A, Buck D, et al. Development, validation, and evaluation of the PBC-40, a disease specific health related quality of life measure for primary biliary cirrhosis. Gut. 2005;54(11):1622–9. 10.1136/gut.2005.065862. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Newton JL, Pairman J, Sutcliffe K, Wilton K, Jones DE. A predictive model for fatigue and its etiologic associations in primary biliary cirrhosis. Clin Gastroenterol Hepatol. 2008;6(2):228–33. 10.1016/j.cgh.2007.11.021. Medline: [DOI] [PubMed] [Google Scholar]
- 29.Doane DP, Seward LE. Measuring skewness: a forgotten statistic? J Stat Educ. 2011;19(2):1–18. 10.1080/10691898.2011.11889611. [DOI] [Google Scholar]
- 30.Razali NM, Wah YB. Power comparisons of Shapiro-Wilk, Kolmogorov-Smirnov, Lilliefors and Anderson-Darling tests [Internet]. J Stat Model and Anal. 2011. [cited 2018 Nov 8];2(1):21–33. Available from: http://www.de.ufpb.br/~ulisses/disciplinas/normality_tests_comparison.pdf [Google Scholar]
- 31.Eron J, Jr., Yeni P, Gathe J Jr, et al.; KLEAN study team. The KLEAN study of fosamprenavir-ritonavir versus lopinavir-ritonavir, each in combination with abacavir-lamivudine, for initial treatment of HIV infection over 48 weeks: a randomised non-inferiority trial. Lancet. 2006;368(9534):476–82. 10.1016/S0140-6736(06)69155-1. Medline:. Erratum in: Lancet. 2006;368(9543): 1238. Corrected at: . [DOI] [PubMed] [Google Scholar]
- 32.Milinkovic A, Benn P, Arenas-Pinto A, et al.; MiPEP Trial Team. Randomized controlled trial of the tolerability and completion of maraviroc compared with Kaletra® in combination with Truvada® for HIV post-exposure prophylaxis (MiPEP Trial). J Antimicrob Chemother. 2017;72(6):1760–8. 10.1093/jac/dkx062. Medline: [DOI] [PubMed] [Google Scholar]
- 33.Walmsley S, Avihingsanon A, Slim J, et al. Gemini: a noninferiority study of saquinavir/ritonavir versus lopinavir/ritonavir as initial HIV-1 therapy in adults. J Acquir Immune Defic Syndr. 2009;50(4):367–74. 10.1097/QAI.0b013e318198a815. Medline: [DOI] [PubMed] [Google Scholar]
- 34.Lytvyak E, Montano-Loza A, Saxinger L, Mason A. Combination anti-retroviral therapy provides reduction in human betaretrovirus load and durable biochemical responses in patients with primary biliary cirrhosis. Hepatology. 2015;62(S1):Section 641. 10.1002/hep.28214. [DOI] [Google Scholar]
- 35.Palella FJ Jr., Delaney KM, Moorman AC, et al.; HIV Outpatient Study Investigators. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection.. N Engl J Med. 1998;338(13):853–60. 10.1056/NEJM199803263381301. Medline: [DOI] [PubMed] [Google Scholar]
- 36.Hayden RT, Gu Z, Ingersoll J, et al. Comparison of droplet digital PCR to real-time PCR for quantitative detection of cytomegalovirus. J Clin Microbiol. 2013;51(2):540–6. 10.1128/JCM.02620-12. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Strain MC, Lada SM, Luong T, et al. Highly precise measurement of HIV DNA by droplet digital PCR. PLoS One. 2013;8(4):e55943. 10.1371/journal.pone.0055943. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Ruelle J, Yfantis V, Duquenne A, Goubau P. Validation of an ultrasensitive digital droplet PCR assay for HIV-2 plasma RNA quantification. J Int AIDS Soc. 2014;17(4 Suppl 3):19675. 10.7448/IAS.17.4.19675. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Langmann P, Hubert C, Heinz W, et al. Safety of long-term lopinavir plasma-levels in patients with liver disease. Eur J Med Res. 2008; 13(5):205–8. Medline: [PubMed] [Google Scholar]
- 40.Verweel G, Burger DM, Sheehan NL, Bergshoeff AS, Warris A, van der Knaap LC, Driessen G, et al. Plasma concentrations of the HIV-protease inhibitor lopinavir are suboptimal in children aged 2 years and below. Antivir Ther. 2007;12(4):453–8. Medline: [PubMed] [Google Scholar]
- 41.Marcelin AG, Cohen-Codar I, King MS, et al. Virological and pharmacological parameters predicting the response to lopinavir-ritonavir in heavily protease inhibitor-experienced patients. Antimicrob Agents Chemother 2005;49(5):1720–6. 10.1128/AAC.49.5.1720-1726.2005. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]