Benefits of prolonged-release pirfenidone plus standard of care treatment in patients with advanced liver fibrosis: PROMETEO study

Established advanced liver fibrosis (ALF) is considered a progressive disease, induced by several etiologies, with no definitive treatment approved by regulatory authorities. It is well established that, unless the injury is blocked or neutralized, most patients suffer complications due to liver failure and increased risk of hepatocellular carcinoma.

Evidence that liver fibrosis regresses to some extent is well-documented, both in experimental models [1, 2] and human liver disease [3, 4]. For fibrosis to regress, the underlying etiology must be treated [5]. Treatment of the underlying disease and concomitant fibrosis regression is associated with improved clinical outcomes. Fibrosis and even cirrhosis regression is characteristic of virtually all forms of liver disease. As yet, however, there is no established treatment that specifically targets liver fibrosis itself [6].

PFD is an antifibrotic drug [7] approved and with granted commercial authorization in Europe, Japan, the USA, Canada, and Mexico for the treatment of idiopathic pulmonary fibrosis (IPF). In Mexico, it has also been granted approval to treat ALF. PFD provides meaningful clinical effects on reductions in the decrease in forced vital capacity (FVC), 6-min walk distance test (6MWT), and mortality, and has improved the progression-free survival of IPF patients with mild-to-moderate disease [8]. PFD (1800–2400 mg daily) taken with food [9], and under adherence to specific access schemes [10] is well tolerated, the most common side effects being gastrointestinal discomfort and photosensitivity. PFD has a favorable benefit–risk profile and represents a suitable treatment option for patients with mild-to-moderate IPF.

This has made it an appealing candidate for other fibrotic diseases [11]. However, concerns about PFD’s potential toxicity must be addressed in diseases that have less ominous prognosis/survival rates. This is particularly pertinent for patients with liver disease, since PFD is mostly metabolized in the liver by cytochrome P450. Armendariz-Borunda et al. have shown that patients with mild ALF treated with standard-release PFD showed a reduction in inflammation and fibrosis of 30% at 12 months [12] and 67% at 24 months without significant deleterious side effects [13].

Our study rationale was to evaluate a drug with well-known effects on multiple organs and tissues, and particularly based on the knowledge that PFD inhibits TGF-β1-induced over-expression of collagen type I [11], a common pathway involved in the pathogenesis of liver cirrhosis from different etiologies.

Our study aimed to determine whether therapy with a new prolonged-release formulation of PFD (PR-PFD), specifically designed to reduce toxicity and maintain constant serum blood levels for longer duration, in combination with standard of care therapy, would result in reducing liver fibrosis and offer an even more favorable benefit–risk profile in patients with ALF.

This was a real-life, multicenter, open-label, proof-of-concept trial to determine the safety and efficacy of 12 months of treatment with PR-PFD in combination with standard of care treatment in adult patients with chronic liver disease whose fibrosis continued to progress despite abstaining from alcohol (ALD), achieving 1 year sustained virologic response (VHC), or otherwise maintaining stable disease (NAFLD, AIH) including healthy life style recommendations. Fibrosis status was determined on the basis of clinical, biochemical, ultrasound and endoscopic findings compatible with chronic damage and at least two other non-invasive methods that confirmed ALF (F3–F4).

A flowchart of the total study population is shown in Fig. 1. Only 122 participants with ALF (F3-F4) completed study-medication for at least 12 months. As expected in an open, real-life trial, some patients (n = 71) did not provide the fibrosis evaluation methods. However, an intention-to-treat (ITT) analysis showed no statistically significant differences between patients included and those who were not included (baseline data in Table 1).

This first-ever, real-life, multicenter, open-label study shows a significant effect of PR-PFD in reducing ALF. Significant beneficial effects were also noted in several clinical, biochemical, and molecular parameters.

Treatment with a prolonged-release formulation of PR-PFD, in combination with a standard of care regimen, resulted in either a FRP or a FSP in the vast majority (35% and 46.7%, respectively), confirming PR-PFD’s significant antifibrotic effects in ALF and extending previous findings [24, 25]. A remarkable difference was observed when compared to only 4.1% FRP seen in the control group, receiving the standard of care regimen.

While some clinicians used to consider cirrhosis irreversible, there are well-documented cases of reversibility of ALF [3], largely correlated with effective treatment of the particular etiology [4, 5]. However, in some patients, fibrosis progresses despite effective treatment. Accordingly, there is a need for ongoing research for treatments directed at fibrosis itself.

This has stimulated evaluation of new antifibrotic drugs [6] in hepatic disorders. Promising studies, like those of Armendariz-Borunda et al. [12, 13] show reduction in both inflammation and fibrosis with negligible side effects. Nonetheless, more rigorous studies concerning PFD’s safety are mandatory, particularly in ALF and given that drug toxicity could increase due to liver impairment.

In our study, PR-PFD was safe and generally well tolerated. Treatment adherence was high, partially explained by patient motivation and a friendly dosing schedule. The type and frequency of AE were consistent with the known safety profile of PFD [8, 10, 11], AEs were typically mild or moderate, and none led to permanent treatment discontinuation. No deaths were detected in the 12-month treatment population and few overall deaths in the whole cohort of eligible population, only half of those liver related.

Our data on blood plasma levels allowed us to evaluate adherence to treatment and better understand the relationship between plasma levels and fibrosis response. In addition to confirming detectable plasma levels in all participants, we identified higher and significant PFD-plasma levels in FRP patients than in FPP patients, suggesting better drug impregnation in this subgroup of responders. Interestingly, delta values between fasting and 2-h post-prandial levels were similar in FRP patients but different in FPP patients, indicating a possible metabolic effect at the hepatic cytochrome level. In addition, this is the first study that provides information of medication levels in a cirrhotic population.

Because the presence of food significantly reduces the extent of absorption of PFD [20], all participants were instructed to take medication after food to reduce known side effects. The plasma levels we found do not point to an inhibitory effect of absorption in FFP patients, but rather a different hepatic metabolism. Around 60% of PFD is bound to plasma proteins, especially to albumin [11], which was found in similar levels in FPP and FRP patients. Since up to 50% of the drug is metabolized by the hepatic CYP1A2 enzyme system to yield 5-carboxy pirfenidone, the inactive metabolite, it is possible that FRP patients have a different enzymatic activity. It is important to keep in mind that our patients were not using fluvoxamine, amiodarone, or propafenone, considered inhibitors of the enzymatic activity of CYP1A2. Similarly, our cohort was not using ciprofloxacin, another CYP1A2 inhibitor. Moderate inducers of CYP1A2, such as tobacco smoking or omeprazole, can reduce the circulating plasma levels of the drug [21]. While none of our patients smoked tobacco, some were taking PPIs. Nevertheless, there was no difference between the FPP and FRP subgroups.

Armendariz-Borunda et al. [22], have suggested that the effectiveness of PFD could be influenced by inherited genetic polymorphisms that increase the risk of developing ALF in some patients. The different PK plasma levels in our study confirm that possibility. Similar clinical trials must consider PFD plasma levels to improve our understanding.

It is assumed that cytokines play central roles in the progression from chronic liver injury to fibrosis/cirrhosis and several pro-inflammatory cytokines (such as IL6, and TNFα) correlate with disease severity [23]. In addition, evidence suggests that hepatic stellate cells express the highest levels of endothelin receptors that may contribute to hepatic sinusoidal tone and therefore portal hypertension [24]. In our study, TGFβ1 serum levels were significantly decreased in PR-PFD-treated patients compared to untreated patients. We also found significant reductions in IL6 and endothelin levels. This suggests that cytokine changes may help predict liver fibrosis evolution.

One of the remarkable facts of our study was the inclusion of patients with different ALF etiologies, as it happens in real life. Our population was recruited from both academic hospitals and ambulatory care units in order to mimic the general population profile, which could increase the external validity of our findings. We understand that fibrosis progression may behave differently in relation to etiology of the liver disease, particularly in patients with AIH. However, according to Hartl et al. [25], transient elastography has been shown to be a reliable tool to monitor disease course in AHI.

In relation to higher relative frequency of HCV among controls (40.5%) compared to PFD-treated population (22%), described in Table 1, we consider this finding is explained because our selection criteria required patients with HCV-treatment and at least 1 year of sustained viral response.

A potential limitation is the absence of liver biopsy, a method with many drawbacks, such as sampling error, cost, and risk of complications [26]. This is mitigated, first, because that, too, mimics the population typical of ambulatory care where non-invasive methods of estimating liver fibrosis are frequently used [27]. Second, non-invasive assessments have been developed and adopted in international management guidelines [15]. LSM by different elastographic methods or magnetic resonance has been found to have an excellent correlation with biopsy findings and is increasingly used in clinical trials. Similarly, FT has been extensively evaluated in patients with chronic liver disease due to diverse etiologies and also has a high correlation biopsy findings, particularly in patients with ALF. According to Lau-Corona et al. [28], FT could enhance our ability to assess differences in fibrosis scores in clinical studies and improve our understanding of fibrosis progression. Recently, Mauro et al. [29], evaluated the value of portal pressure, liver stiffness, and enhanced liver fibrosis score measurements to predict fibrosis regression according to paired liver biopsies before and after sustained viral response (SVR) in recurrent HCV patients. They concluded that the dynamic changes in LSM accurately predict the presence of ALF and clinically significant portal pressure 1 year after SVR and thus can be used as reliable monitoring strategies. Chalasani et al. [30], confirm the utility of non-invasive markers in ALF patients by showing that longitudinal changes in non-invasive measures of fibrosis correlate with improvements in histologic fibrosis and therefore can serve as surrogate end points in clinical trials. Third, we used two validated, non-invasive methods.

It is worth mentioning that we observed an 11% reduction in body weight throughout the duration of the study in the pirfenidone-treated group. Since diet and exercise are known factors to influence disease progression/regression, and even portal pressure reduction, we consider that controlling these factors should be part of the commitments of the standard of care.

Another important finding is significant improvements in liver function and liver enzymes. Child–Pugh is considered a strong bedside prognostic indicator in advanced liver disease, often used with MELD to determine need and priority for transplant [18]. Here, the vast majority improved both scores or remained steady, which correlates with an increase in serum albumin, another clinical prognostic indicator.

Contrary to what might be expected regarding possible liver toxicity, no patients’ AST/ALT levels increased X5 upper limit of normal (ULN). Rather, we identified a significant decrease in 40.6% and 43.3% in ALT and AST serum levels, demonstrating a favorable safety profile. In Angulo et al. study [31], which reports a high proportion of AE in 24 primary sclerosing cholangitis patients, treatment was with standard-release PFD and at a much higher daily dose (2400–3600 mg). Key strategies to prevent and manage common PFD AEs have been described to maximize adherence and prevent AEs [10].

Regarding the suitability of the PFD formulation that we used in our study, PR-PFD offers a lower C_max and longer T_max and half-life pharmacokinetic profile compared to standard-release PFD [20].

PR-PFD association with lower PFD C_max values (Poo JL, et al. Pharmacokinetics of the antifibrotic drug pirfenidone in Child–Pugh A and B cirrhotic patients compared to healthy age-matched controls. Journal of Hepatology 2016;64:S213–S424) may explain the excellent tolerability in our study. Also promising is that the self-assessment questionnaires indicated significant QOL effects, mainly described as improvement in energy and reduction in fatigue.

In conclusion, prolonged-release PFD administration in conjunction with standard of care treatment in patients with advanced liver fibrosis provides beneficial effects that warrant additional studies. Our data suggest that inflammation and liver stiffness could be ameliorated by PFD-treatment, a finding needing confirmation in a placebo-controlled clinical trial.