PIVENS Trial: A Plain-English Overview of What It Established

At a glance

• Full title: Pioglitazone, Vitamin E, or Placebo for Nonalcoholic Steatohepatitis
• Registry: NCT00063622 (NASH CRN)
• N: 247 adults with biopsy-confirmed NASH, no diabetes
• Arms: Vitamin E 800 IU/day (n=84) vs. pioglitazone 30 mg/day (n=80) vs. placebo (n=83)
• Duration: 96 weeks of treatment with liver biopsy at baseline and week 96
• Primary endpoint: Improvement in NAFLD Activity Score (NAS) by ≥2 points, with at least 1-point reduction in both lobular inflammation and hepatocyte ballooning, and no worsening of fibrosis
• Key result: Vitamin E met the primary endpoint (43% vs. 19% placebo, p=0.001); pioglitazone did not reach statistical significance on the primary composite (34% vs. 19%, p=0.04 against a prespecified threshold of p<0.025)
• Published: 2010, New England Journal of Medicine

The Question PIVENS Set Out to Answer

By the mid-2000s, clinicians knew that nonalcoholic steatohepatitis (NASH, now called MASH) could progress to cirrhosis, liver failure, and hepatocellular carcinoma. What they lacked was any proven pharmacotherapy. Two candidates had biological rationale: vitamin E, an antioxidant that could counter the oxidative stress driving hepatocyte injury, and pioglitazone, an insulin sensitizer that could reduce the metabolic dysfunction feeding fat accumulation and inflammation.

The NASH Clinical Research Network (CRN), funded by the National Institute of Diabetes and Digestive and Kidney Diseases, designed PIVENS to answer a direct question: does either agent improve the actual tissue-level disease when compared with placebo over two years?

Who Was Enrolled (and Who Was Not)

PIVENS recruited 247 adults aged 18 and older at eight U.S. academic centers. Every participant had biopsy-confirmed NASH with a NAS of 4 or higher and no evidence of other chronic liver diseases. The most consequential enrollment decision was the exclusion of patients with diabetes. The investigators reasoned that pioglitazone is already an approved diabetes drug, and including diabetic patients would confound interpretation of liver-specific benefit versus glycemic improvement. This means PIVENS results cannot be directly extrapolated to the large population of patients who have both type 2 diabetes and NASH.

Baseline characteristics were well matched across the three arms. The average age was approximately 46 years. Roughly 60% were women, and 83% were white. Mean BMI hovered around 34 kg/m², and average NAS was about 5.3, indicating moderate-to-severe disease. About 29% of participants had bridging fibrosis (stage 3) at entry.

What Each Group Received

Patients were randomized 1:1:1 to one of three arms:

| Arm | Dose | Route | Duration | |---|---|---|---| | Vitamin E | 800 IU/day (RRR-alpha-tocopherol) | Oral | 96 weeks | | Pioglitazone | 30 mg/day | Oral | 96 weeks | | Placebo | Matched capsules | Oral | 96 weeks |

The trial used a double-dummy design so that every patient took both a capsule (vitamin E or its placebo) and a tablet (pioglitazone or its placebo). This preserved blinding. All patients received standard lifestyle counseling, but no structured diet or exercise program was imposed.

How the Primary Endpoint Actually Worked

The primary endpoint was a composite histological outcome, not a single lab value or imaging marker. A central pathology committee, blinded to treatment assignment, scored biopsies at baseline and week 96 using the NAS system. To count as a "responder," a patient needed all three of the following:

1. Total NAS improvement of ≥2 points from baseline
2. At least a 1-point drop in lobular inflammation
3. At least a 1-point drop in hepatocyte ballooning
4. No increase in fibrosis stage

This composite is strict. A patient whose NAS fell by 3 points entirely from reduced steatosis, with no change in ballooning, would not qualify. The design prioritized evidence that the drug was fixing the inflammatory and injury components of NASH, not just reducing fat content.

Because two active agents were tested against one placebo group, the trial used a Hochberg procedure to control the family-wise error rate, setting significance at p<0.025 for each comparison.

The Results in Detail

Primary Endpoint

| Arm | Responders | Rate | p vs. Placebo | |---|---|---|---| | Vitamin E | 36/84 | 43% | 0.001 | | Pioglitazone | 27/80 | 34% | 0.04 | | Placebo | 16/83 | 19% |, |

Vitamin E clearly met the prespecified threshold. Pioglitazone's p-value of 0.04 would be conventionally significant, but the Hochberg correction required p<0.025, so it formally failed the primary endpoint.

Resolution of NASH (Key Secondary)

This secondary endpoint asked: did the biopsy at week 96 no longer meet diagnostic criteria for NASH?

| Arm | Resolution Rate | |---|---| | Vitamin E | 36% | | Pioglitazone | 47% | | Placebo | 21% |

Pioglitazone's 47% resolution rate was the highest of any arm, reaching statistical significance versus placebo (p=0.001). Vitamin E's 36% was also significant (p=0.05). This created an unusual situation: pioglitazone performed better on NASH resolution yet missed the primary composite, while vitamin E won the primary composite but showed lower resolution rates.

Component Scores

| Component | Vitamin E Change | Pioglitazone Change | Placebo Change | |---|---|---|---| | Steatosis (0-3) | −0.7 | −0.8 | −0.3 | | Lobular inflammation (0-3) | −0.7 | −0.5 | −0.2 | | Ballooning (0-2) | −0.5 | −0.4 | −0.2 | | Fibrosis (0-4) | −0.1 | −0.1 | −0.1 | | Total NAS (0-8) | −1.9 | −1.7 | −0.7 |

Both agents reduced steatosis, inflammation, and ballooning compared with placebo. Neither agent improved fibrosis more than placebo over 96 weeks.

Metabolic and Biochemical Outcomes

Serum ALT dropped by a mean of 37 U/L in the vitamin E group and 40 U/L in the pioglitazone arm, versus 20 U/L in placebo. Pioglitazone also reduced insulin resistance (measured by HOMA-IR), which vitamin E did not. However, mean body weight increased by 4.7 kg in the pioglitazone arm compared with 0.4 kg with vitamin E and 0.7 kg with placebo.

Safety Signals and Weight Gain

Vitamin E was generally well tolerated in this trial. There were no significant differences in serious adverse events between vitamin E and placebo over 96 weeks.

Pioglitazone carried a clear weight gain penalty. The average 4.7 kg increase is consistent with the thiazolidinedione class effect on adipose tissue expansion. Peripheral edema was more common in the pioglitazone group, though heart failure events were not observed. Post-trial concerns about pioglitazone's bladder cancer risk emerged from other datasets, though the FDA's updated label notes the evidence remains inconclusive.

For vitamin E, longer-term observational data raised questions about a potential association between high-dose supplementation and all-cause mortality or prostate cancer risk. A meta-analysis by Miller et al. (2005) suggested a small increase in all-cause mortality at doses above 400 IU/day, and the SELECT trial reported a 17% relative increase in prostate cancer incidence with vitamin E supplementation. These findings are not definitive, but they inform the risk-benefit conversation for long-term use.

Why the Primary Endpoint Result Is More Nuanced Than It Looks

The headline "vitamin E won, pioglitazone lost" oversimplifies what happened. Pioglitazone's p-value of 0.04 against the 0.025 threshold reflects a statistical design choice, not a biological failure. On nearly every histological and biochemical measure, pioglitazone performed between vitamin E and placebo, or on par with vitamin E. Its NASH resolution rate of 47% actually exceeded vitamin E's 36%.

The AASLD practice guidance updated in 2023 acknowledges both agents as options, recommending vitamin E (800 IU/day) for non-diabetic adults with biopsy-proven NASH and noting pioglitazone as an option for patients with or without type 2 diabetes. The guidance explicitly states that pioglitazone can be considered even in patients without diabetes, citing PIVENS as supporting evidence despite the technically negative primary endpoint.

Limitations the Authors Acknowledged

The original publication and the NASH CRN group were transparent about several constraints:

• No diabetic patients. The largest at-risk population was excluded, limiting generalizability.
• 96-week duration. NASH progresses over decades. Two years cannot capture effects on hard outcomes like cirrhosis, transplant need, or death.
• No fibrosis improvement. Neither drug moved the fibrosis needle, which is the histological feature most strongly linked to long-term liver outcomes.
• Rebound after discontinuation. Follow-up data from the NASH CRN suggested that ALT levels rose again after treatment was stopped, raising questions about whether indefinite therapy would be needed.
• Biopsy-driven endpoints. Liver biopsy carries sampling variability. A single 1/50,000th slice of the liver may not represent the whole organ, and repeated biopsies introduce procedural risk.

What PIVENS Means for Practice Today

PIVENS was published in 2010. In 2024, resmetirom (Rezdiffra) became the first FDA-approved drug for NASH/MASH with moderate-to-advanced fibrosis. For the 14 years between PIVENS and that approval, vitamin E and pioglitazone were the only pharmacological options with RCT support, and they remain relevant for patients who do not have the fibrosis stage or insurance coverage required for resmetirom.

Vitamin E remains a first-line recommendation for non-diabetic NASH patients willing to accept the uncertain long-term safety profile. Pioglitazone occupies a broader niche because it can be used in diabetic patients (an approved indication for glucose control that also yields liver benefit), though weight gain and fluid retention limit its acceptability. Neither drug has been shown to reverse fibrosis in a randomized trial, which is the unmet need that newer agents like resmetirom, semaglutide, and lanifibranor are designed to address.

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. 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
3. Miller ER 3rd, Pastor-Barriuso R, Dalal D, et al. Meta-analysis: high-dosage vitamin E supplementation may increase all-cause mortality. Ann Intern Med. 2005;142(1):37-46. PubMed
4. 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
5. FDA. Rezdiffra (resmetirom) prescribing information. 2024. FDA Label
6. FDA. Actos (pioglitazone) prescribing information. 2011. FDA Label