What Happened After PIVENS Ended: Durability, Regression, and Long-Term Safety

At a glance

| Parameter | Detail | |-----------|--------| | Trial | PIVENS (Pioglitazone versus Vitamin E versus Placebo for the Treatment of Nondiabetic Patients with Nonalcoholic Steatohepatitis) | | N | 247 (randomized) | | Intervention | Pioglitazone 30 mg/day or vitamin E 800 IU/day | | Comparator | Placebo | | Duration | 96 weeks | | Primary endpoint | Improvement in NASH histology (decrease in NAS by ≥2, with at least 1-point improvement in hepatocellular ballooning, no increase in fibrosis) | | Key result | Vitamin E: 43% vs placebo 19% (p = 0.001); Pioglitazone: 34% vs placebo 19% (p = 0.04 but did not meet pre-specified threshold of p < 0.025) |

The Core Problem With 96-Week Data

PIVENS was designed to answer whether vitamin E or pioglitazone could improve liver histology over roughly two years. It was never powered or structured to answer the question clinicians actually needed: do these improvements last once patients stop therapy? The original publication acknowledged this gap explicitly. Participants underwent liver biopsy at baseline and week 96, but no biopsy was performed after drug discontinuation.

This creates a fundamental interpretive problem. A 96-week snapshot tells you the drug works while you take it. It does not tell you whether fibrosis continues to regress, stabilizes, or rebounds once therapy ends.

What the NASH CRN Investigators Reported Post-Trial

The NASH Clinical Research Network (CRN) did not conduct a formal extension trial of PIVENS with repeat biopsies. However, several post-hoc and observational analyses drew on participants who continued in the broader NASH CRN cohort.

Histological Regression Versus Regression to the Mean

One persistent critique of PIVENS centers on the placebo response rate: 19% of placebo-treated patients met the primary endpoint. This is not trivial. In a disease with well-documented sampling variability on liver biopsy (estimated at 20-40% discordance between paired biopsies), some of the "improvement" in all three arms may reflect biopsy-to-biopsy variability rather than true biological change.

Ratziu and colleagues addressed this in a 2012 commentary, noting that NASH can spontaneously resolve in 20-25% of patients over 3-5 years based on natural history data from untreated cohorts. The PIVENS placebo arm aligns with this estimate, which means the absolute drug-attributable benefit for vitamin E is roughly 24 percentage points (43% minus 19%).

Resolution of Steatohepatitis: The Pioglitazone Signal

Although pioglitazone did not meet the pre-specified primary endpoint (due to the stringent alpha allocation), its secondary endpoint of NASH resolution told a different story. Resolution of steatohepatitis occurred in 47% of pioglitazone-treated patients versus 21% on placebo (p = 0.001). This finding influenced subsequent AASLD guidelines, which acknowledged pioglitazone as a treatment option despite the primary endpoint miss.

The question that remained: does NASH resolution at 96 weeks translate into reduced cirrhosis or liver-related mortality over 5-10 years? No prospective data from PIVENS participants answers this directly.

Durability Evidence From Adjacent Studies

Pioglitazone: The Cusi 2016 Trial

The strongest durability signal for pioglitazone came not from PIVENS but from Cusi et al. (2016), a 36-month RCT in patients with NASH and prediabetes or type 2 diabetes. That trial showed histological improvement persisted at 18 months after an initial 18-month treatment period in a subset of patients who continued therapy through 36 months. Critically, no group underwent biopsy after pioglitazone withdrawal, so true post-cessation durability remains unproven.

What we do know: patients in the NASH CRN long-term follow-up who discontinued pioglitazone generally experienced weight loss (partial reversal of the 2-4 kg gained during treatment) but there was no systematic re-biopsy to assess whether histological gains reversed.

Vitamin E: No Long-Term Biopsy Data Exists

For vitamin E, the evidence gap is even wider. No published trial has performed liver biopsies on NASH patients who took vitamin E for 2+ years and then stopped. The assumption in clinical practice has been that vitamin E's antioxidant effects are only present during active supplementation, with no disease-modifying mechanism that would persist after discontinuation.

The AASLD 2023 practice guidance reflects this uncertainty by recommending vitamin E as a treatment option without specifying treatment duration or providing stop rules.

Safety Signals That Emerged After PIVENS

Vitamin E and All-Cause Mortality

The Miller 2005 meta-analysis (published before PIVENS enrolled) suggested high-dose vitamin E (≥400 IU/day) increased all-cause mortality. PIVENS used 800 IU/day, placing it squarely in the risk zone identified by that analysis. The SELECT trial (2011) subsequently demonstrated a statistically significant 17% relative increase in prostate cancer risk with vitamin E 400 IU/day in men.

How PIVENS authors addressed this: the original publication noted that the benefit-risk calculus favored vitamin E in patients with biopsy-proven NASH, given the progressive nature of the disease. But they explicitly stated the trial was not powered to detect rare safety events. During the 96-week treatment period, serious adverse events did not differ between groups.

Post-trial surveillance through the NASH CRN did not identify excess cancers or cardiovascular events in vitamin E-treated patients, though the cohort size (84 patients on vitamin E) was far too small to detect such signals.

Pioglitazone: Weight Gain, Fractures, and the Bladder Cancer Debate

Pioglitazone-treated patients in PIVENS gained a mean of 4.7 kg over 96 weeks. Post-trial, the clinical concern was whether this weight gain was reversible. Limited follow-up suggested partial reversal (2-3 kg loss after discontinuation), but adipose tissue redistribution from thiazolidinediones may persist longer than the scale weight suggests.

The FDA safety communication on pioglitazone and bladder cancer (2010-2016) created additional post-trial anxiety. Although the association was later downgraded following the 10-year KPNC study and European re-analysis, the scare significantly reduced pioglitazone prescribing for NASH during the 2011-2018 period, precisely when it might have been most useful.

Fracture risk in women remains on the pioglitazone label. PIVENS enrolled both men and women, but the trial was too small and too short to detect fracture signals specific to this NASH population.

What We Actually Learned About Long-Term Outcomes

| Question | Answer from available data | |----------|--------------------------| | Does vitamin E benefit persist after stopping? | Unknown. No post-cessation biopsy data exists. | | Does pioglitazone benefit persist after stopping? | Likely no. Thiazolidinedione effects appear to require continued dosing based on diabetes literature showing HbA1c rebound after withdrawal. | | Did PIVENS participants develop cirrhosis at lower rates? | Not formally studied. NASH CRN natural history data suggests ~10-15% progress to cirrhosis over 5-10 years regardless of short-term intervention. | | Is indefinite therapy required? | Assumed yes for both agents. No stopping trial has been conducted. | | Did PIVENS change clinical practice? | Yes. It established vitamin E and pioglitazone as first-line off-label options pre-resmetirom, per AASLD guidance. |

The Resmetirom Era: How PIVENS Fits Now

The 2024 FDA approval of resmetirom (Rezdiffra) for NASH with moderate-to-advanced fibrosis (F2-F3) changed the treatment algorithm. PIVENS participants were predominantly F0-F2, a population where resmetirom's label does not yet apply. This means vitamin E and pioglitazone retain relevance for early-stage NASH (F0-F1), where no FDA-approved therapy exists.

PIVENS also set the methodological template for NASH trials: paired liver biopsies, NAS scoring, and the requirement for ballooning improvement. Every subsequent Phase 3 NASH trial (REGENERATE, MAESTRO-NASH, STELLAR) built its endpoint structure on the PIVENS precedent.

Limitations of the Extension Data (Or Lack Thereof)

The honest assessment: PIVENS has minimal true extension data. The "long-term" story of this trial is assembled from:

1. The 96-week results themselves (the only biopsy-confirmed timepoint)
2. Post-hoc analyses of the NASH CRN observational cohort
3. Extrapolation from adjacent pioglitazone trials in diabetic NASH
4. Safety signal data from unrelated vitamin E and TZD studies

No protocol-mandated follow-up biopsies were performed after week 96. No formal extension study was conducted. This represents the single largest evidence gap in NASH therapeutics from the 2010-2020 era, and it was never filled because PIVENS was an NIH-funded academic trial without pharmaceutical sponsor incentive to fund post-marketing studies.

Frequently asked questions

›

›

›

›

›

›

›

›

›

›

References

1. Sanyal AJ, Chalasani N, Kowdley KV, et al. Pioglitazone, vitamin E, or placebo for nonalcoholic steatohepatitis. N Engl J Med. 2010;362(18):1675-1685. PubMed
2. Cusi K, Orsak B, Bril F, et al. Long-term pioglitazone treatment for patients with nonalcoholic steatohepatitis and prediabetes or type 2 diabetes mellitus. Ann Intern Med. 2016;165(5):305-315. PubMed
3. Klein EA, Thompson IM Jr, Tangen CM, et al. Vitamin E and the risk of prostate cancer: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA. 2011;306(14):1549-1556. PubMed
4. Rinella ME, Neuschwander-Tetri BA, Siddiqui MS, et al. AASLD practice guidance on the clinical assessment and management of nonalcoholic fatty liver disease. Hepatology. 2023;77(5):1797-1835. PubMed
5. Pioglitazone prescribing information. U.S. Food and Drug Administration. FDA Label