Many physicians do not consider liver disease and liver cancer classic complications of obesity, type 2 diabetes, or metabolic syndrome, but they should.

Research findings over the past decade offer substantial evidence for links between obesity, diabetes, or metabolic syndrome and the earliest hepatic manifestation of these derangements: nonalcoholic fatty liver disease (NAFLD). Equally compelling links tie obesity, diabetes, and metabolic syndrome to more advanced liver pathology: nonalcoholic steatohepatitis (NASH), cirrhosis, and liver cancer, especially hepatocellular carcinoma (HCC).

Although the link between obesity, diabetes, or metabolic syndrome and NASH or liver cancer is not yet strong enough to justify major changes in disease surveillance or management, the link between these metabolic disorders and NAFLD is powerful and common enough to warrant routinely considering these patients as having NAFLD, say experts. And if NAFLD is found, the next step is deciding if a patient is the right candidate for NASH or cirrhosis assessment; and if those disorders develop, cancer screening follows.

A new dimension of obesity and diabetes morbidity

"For decades, attention to the patient with type 2 diabetes focused on the control of hyperglycemia and of risk factors associated with macrovascular disease. The epidemic of obesity now presents endocrinologists with new challenges. Among them is the need to identify early complications related to obesity in the setting of type 2 diabetes. NAFLD is a common complication of patients with type 2 diabetes that ... does not fit into the traditional realm of diabetic complications," Dr. Romina Lomonaco and Dr. Kenneth Cusi wrote in a recently published book chapter ("Evidence-based Management of Diabetes," chapter 21; TFM Publishing, 2012).

Not until recently has NAFLD been recognized as another common complication of patients with type 2 diabetes that requires special attention. NAFLD’s low profile as a complication of obesity and diabetes contrasts with its ubiquity. About 70% of patients with type 2 diabetes have NAFLD (compared with about 20% of all American adults), and perhaps up to 90% of morbidly obese patients have NAFLD. The prevalence of impaired fasting glucose and of newly diagnosed type 2 diabetes is about threefold higher in patients with NAFLD than in those without liver disease.

"Insulin resistance and obesity are probably the biggest factors" causing NAFLD, said Dr. Cusi, professor of medicine and chief of adult endocrinology, diabetes, and metabolism at the University of Florida in Gainesville. Moreover, "diabetes will worsen NAFLD, producing more fibrosis and an increased rate of cirrhosis," he said in an interview.

That’s significant because it is NAFLD progression that poses the biggest risk. NAFLD severity can range from mild, early-stage disease in an asymptomatic patient with normal liver enzyme levels to the development of inflammation –NASH – which can cause liver injury and fibrosis, lead to cirrhosis, and set up progression to organ failure or development of HCC or other liver cancer.

Overall, about 40% of patients with NAFLD progress to NASH, but both obesity and diabetes ratchet up NAFLD progression, and so roughly half of all patients with diabetes have NASH. Patients with type 2 diabetes also have a two- to fourfold increased risk of developing advanced liver disease, cirrhosis, and HCC compared with people without diabetes. "About 15% of NASH patients develop cirrhosis, and a significant percent also develop cancer," Dr. Cusi said.

"NASH represents the hepatic manifestation of the metabolic syndrome, a constellation of abdominal obesity, hypertension, diabetes, and dyslipidemia. It is projected that 25 million Americans will develop NASH by 2025, with 20% progressing to cirrhosis, hepatocellular carcinoma, or both, that may require liver transplantation," wrote Dr. Vatche G. Agopian and his associates from the Dumont-University of California, Los Angeles (UCLA), Transplant and Liver Cancer Center in a recent report (Ann. Surg. 2012;256:624-33).

From 2001 to 2009, the nationwide frequency of NASH as the primary indication for liver transplantation rose from 1% to 10%, with NASH becoming the third most common U.S. indication for liver transplantation (Gastroenterology 2011;141:1249-53). The UCLA surgeons reviewed their experience with 1,294 patients who underwent primary liver transplantation at their center between January 2002 and August 2011, and found 136 patients (11%) who met NASH criteria. But during the 10-year period studied, NASH as the trigger for liver transplant soared from 3% of transplants in 2002 to 19% in 2011, a jump that by 2011 made NASH the second most common cause for liver transplant at UCLA, trailing only hepatitis C virus. In fact, NASH "is poised to surpass hepatitis C as the leading indication in the next 10-20 years," wrote Dr. Agopian, a liver surgeon, and Dr. Busuttil, professor and chief of liver and pancreatic transplantation at UCLA, and their associates (Ann. Surg. 2012;256:624-33).

In their report, Dr. Agopian and Dr. Busuttil called the current surge in liver transplants for patients with NASH "the new epidemic."

"The future of [liver] transplantation is here with these patients who have nonalcoholic steatohepatitis and subsequent cirrhosis," commented Dr. John P. Roberts, professor and chief of transplant surgery at the University of California, San Francisco. "Currently, there are about 6,000 [liver] transplants [per year] in the United States. Half of those are done for hepatitis C. In the overall population of the United States, 1.3% have hepatitis C, and that provides about half of liver transplant patients. Twelve percent of the U.S. population have nonalcoholic steatohepatitis, a 10-fold increase over the percentage of the population with hepatitis C. Due to the kinetics of the hepatitis C epidemic, we are going to see a falloff in the number of patients with hepatitis C eligible for transplantation in the next 10 years. [Patients with NASH] are going to replace them, potentially by 10-fold," said Dr. Roberts, who commented on the report by Dr. Agopian and Dr. Busuttil on the UCLA experience during the 2012 annual meeting of the American Surgical Association in San Francisco.

The NAFLD, NASH, and HCC connections

"The link between obesity, NASH, cirrhosis, and HCC is very strong" said Dr. Stephen H. Caldwell, professor of medicine and director of hepatology at the University of Virginia in Charlottesville.

"What remains unknown is whether NASH and hepatic scar formation are essential to cause cancer, or can carcinomas arise in a noncirrhotic, non-NASH fatty liver? Scar formation itself is a carcinogenic process, especially when it progresses to stage 3 – bridging fibrosis – or to stage 4," when cirrhosis occurs.

"It’s difficult to justify screening all patients with a fatty liver; that would be a huge undertaking," Dr. Caldwell said in an interview. "The more important clinical message is to consider whether a patient has NASH, but that is hard to diagnose without a liver biopsy."

So far, no markers have been unquestionably accurate for diagnosing NASH. Any patient who is obese or has metabolic syndrome should be considered for NASH, said Dr. Caldwell. Signs of more advanced liver injury include cirrhosis or portal hypertension. Other, more subtle signs include spider angiomas, reddening of the palms, declining platelet counts, or a family history of liver disease. Any of these could be a reason to look for NASH, he said.

Last year, guidelines issued by the American Association for the Study of Liver Diseases (AASLD), the American College of Gastroenterology, and the American Gastroenterological Association recommended against routinely testing for NAFLD, even among patients in diabetes or obesity clinics. Evidence was lacking for routine screening, even of high-risk patients, the guidelines said, with no data on cost effectiveness and uncertainties about diagnostic tests and treatment options (Hepatology 2012;55:2005-32).

But the guidelines do call for targeted assessment of NAFLD, and targeting NASH workups for selected NAFLD patients. The guidelines recommend ruling out all other possible etiologies and establishing NAFLD by histology or imaging. When a patient is diagnosed with NAFLD, the guidelines say that "as the metabolic syndrome predicts the presence of steatohepatitis in patients with NAFLD, its presence can be used to target patients for liver biopsy." The 2012 guidelines also highlighted the NAFLD Fibrosis Score (Hepatology 2007;45:846-54) as another useful tool to identify NAFLD patients at increased risk for NASH or cirrhosis. The guidelines called the plasma biomarker cytokeretin-18 "promising," but cautioned that it was "premature to recommend in routine clinical practice."

Major issues for patients who develop NASH are their risk for cirrhosis and liver failure, as well as that for liver cancer. Although the case already exists for obesity, diabetes, and metabolic syndrome as factors leading to NAFLD and NASH, evidence also links each of these three conditions to an increased rate of HCC and other liver cancer, such as cholangiocarcinoma.

"The evidence supports both an independent role for obesity, insulin resistance, and diabetes, as well as boosting the risk from other major risk factors such as hepatitis. The missing evidence is it has not been shown that treatment of diabetes or weight loss can reduce the risk of liver cancer," said Dr. Hashem B. El-Serag, professor and chief of gastroenterology and hepatology at the Baylor College of Medicine in Houston. "Screening for fatty liver by liver enzymes and ultrasound is probably a prudent first step" for obese or insulin-resistant patients, noted Dr. El-Serag. But surveillance for HCC by twice-annual ultrasound exams is only for patients with demonstrated advanced fibrosis or cirrhosis, he said in an interview.

"We currently recommend that anyone with NAFLD cirrhosis or cirrhosis of unknown etiology who is also obese or had diabetes should receive routine HCC surveillance," said Dr. György Baffy, chief of gastroenterology at the VA Boston Healthcare System. He predicts that "we may soon reach a general conclusion that people with morbid obesity (a body mass index of greater that 40 kg/m²) and poorly controlled diabetes should be considered for liver cancer surveillance even without clear evidence for cirrhosis," he said in an interview. But in general, "HCC occurrence in noncirrhotic liver is so low that surveillance would be rather inefficient."

Despite that, Dr. Baffy admits that the connection between diabetes and HCC may go beyond cirrhosis. "Up to half of all HCC may develop in noncirrhotic livers," he wrote in a recent editorial (Am. J. Gastroenterol. 2012;107:53-5). "It is more difficult to determine the need for HCC surveillance in diabetic patients with noncirrhotic liver or with no established liver disease."

To avoid missing a diagnosis of HCC, Dr. Baffy suggested awareness of the risk factors for advanced background liver disease and for HCC in patients with diabetes: male sex, older age, morbid obesity, poorly controlled and long-standing disease, and coexisting hepatitis C.

"For now, cirrhosis remains the primary indication for implementing HCC surveillance," but the new findings on liver cancer developing in liver disease associated with obesity and diabetes so far provide insufficient evidence to warrant any firm screening recommendations for these patients, Dr. Baffy wrote in another recent article along with Dr. Caldwell and Dr. Elizabeth M. Brunt (J. Hepatology 2012;56:1384-91). "The greater dilemma comes from new evidence that HCC may complicate NAFLD when fibrosis is mild or absent. Observations that diabetes may increase the risk of HCC regardless of the presence of cirrhosis remain a major concern for the 26 million Americans estimated to have diabetes or prediabetes," they wrote. "We may need to contemplate a paradigm shift in liver cancer surveillance, but for now at least, HCC appears to be a rare complication of NAFLD in the complete absence of fibrosis."

In addition, the value of regular cancer surveillance, even in patients with cirrhosis, remains uncertain, just as surveillance for breast cancer and prostate cancer has come under similar criticism. "It gets a little shaky when you look for evidence that [HCC] surveillance led to prolonged survival," said Dr. Caldwell. "You have all the same controversy as breast cancer, but surveillance is even less established for HCC."

Diabetes also linked to HCC spread

Once hepatocellular carcinoma forms in a patient with diabetes, the cancer may act more aggressively, according to studies from the University of Rochester (N.Y.).

A review of 265 consecutive patients treated for hepatocellular carcinoma (HCC) at Rochester’s Wilmot Cancer Center identified 91 (34%) with diabetes at the time of HCC diagnosis. Forty-seven of the 265 patients (18%) had distant metastases at the time of diagnosis. A multivariate analysis that controlled for age and etiologic risk factors showed that patients with diabetes were 10 times more likely to have distant metastases at the time of HCC diagnosis, compared with patients without diabetes, Dr. Aram F. Hezel and his associates reported last year (Cancer Investigation 2012;30:698-702). The analysis showed no statistically significant impact of diabetes on survival rate.

In a second analysis they found that patients with newly diagnosed HCC and diabetes were also significantly more likely to have macrovascular invasion by the HCC.

"We don’t treat patients with HCC differently if they have diabetes or obesity, but our findings show an association between diabetes and greater spread of HCC, more invasive cancer," said Dr. Hezel, an oncologist and director of hepatic and pancreatic cancer research at the Wilmot Cancer Center of the University of Rochester (N.Y.). "We don’t know whether we can treat the diabetes and change the behavior of the cancer by having patients under better control. Are cancers different in patients with diabetes or obesity? Do some metabolic states help push a cancer to more invasive behavior?" he asked in an interview.

"We use liver transplant to treat patients with liver cancer. In early stages of liver cancer the tumor is less likely to spread, so liver transplant can be curative. But if there are patients with a greater propensity for cancer spread at an earlier stage" then the efficacy of transplantation needs reassessment, Dr. Hezel said.

Few proven treatments for NAFLD, NASH, and to prevent HCC

Although diagnosing NAFLD is an important step in identifying patients at risk for NASH, cirrhosis, and liver cancer, interventions with proven benefits for NAFLD are limited. No approved drug treatments exist for NAFLD; lifestyle modification is the standard treatment to reduce steatosis and plasma levels of liver aminotransferases. Reductions in liver fat correlate closely with weight loss, Dr. Cusi, Dr. Lomonaco, and their associates said in a recently published analysis of NAFLD (Drugs 2013; Jan. 11 [Epub ahead of print]). A weight loss of 7%-10% has been linked with a roughly 50% drop in liver fat levels in NAFLD patients, they said. But long-term controlled studies are needed to better assess the impact of lifestyle changes on NAFLD and fatty livers.

Pioglitazone received endorsement from the AASLD panel for treating NASH in their 2012 NAFLD management recommendations. The recommendations cautioned that most NASH patients who received pioglitazone treatment in trials did not have diabetes, and that long-term safety and efficacy of pioglitazone in NASH patients are not established.

The AASLD guidelines also call for using vitamin E at a daily dosage of 800 IU, but only for patients with biopsy-proven NASH and no diabetes; the guidelines call it "first line" in this setting. But the guidelines also specifically caution against using vitamin E in patients with NASH and diabetes, patients with NAFLD who have not undergone a liver biopsy, patients with NASH and cirrhosis, and those with cryptogenic cirrhosis. The guidelines also caution against using metformin to treat NASH. No other drug intervention gets guideline endorsement for treating NASH.

"You can say diet and exercise minimize the risk of fatty liver, but beyond that drug therapy is unclear," said Dr. Caldwell. "I think as we see treatment evolve, we’ll see more interest [in treating NAFLD and NASH] by endocrinologists," he predicted.

The intervention picture changes when the goal is preventing liver cancer. "Effective treatment of insulin resistance and hyperinsulinemia may be critical to prevent hepatocarcinogenesis," wrote Dr. Baffy, Dr. Brunt, and Dr. Caldwell in their recent review (J. Hepatology 2012;56:1384-91). "Insulin sensitizing agents in diabetes may reduce the risk of HCC." They especially cited the epidemiologic evidence supporting a role for thiazolidinediones, which were linked to a 70% reduction in HCC incidence among patients with diabetes compared with patients treated with insulin or a sulfonylurea in a case-control study (Cancer 2010;116:1938-46). The same study also showed a similar, 70% reduction in HCC among patients treated with a biguanide like metformin.

"While current guidelines for the management of HCC have no specific recommendations for cases associated with NAFLD, obesity, and diabetes, the use of insulin-sensitizing drugs and avoidance of treatments that contribute to hyperinsulinemia are likely to enhance prevention and improve disease outcomes of HCC," said Dr. Baffy, Dr. Brunt, and Dr. Caldwell.

Similar evidence recently came from other epidemiologic studies that suggest damping down of HCC development in patients treated with a thiazolidinedione or metformin. A report last year that analyzed health records of about 98,000 Taiwan residents found that treatment with a thiazolidinedione or with metformin reduced the rate of HCC in patients with diabetes by about 50% compared with other treatments (Am. J. Gastroenterol. 2012;107:46-52). More evidence supporting protection from metformin against formation of both HCC and a second, less common type of liver cancer, intrahepatic cholangiocarcinoma, came in two studies reported last May at the annual Digestive Disease Week in San Diego.

"Metformin has not proved useful in the therapy of NAFLD, but it is helpful in decreasing the risk of HCC in patients with obesity- or diabetes-associated liver disease. Metformin should be part of antidiabetic management whenever possible," Dr. Baffy said in an interview.

But other experts regard the evidence accumulated so far as too preliminary to guide management. "It is premature to recommend using [metformin or a thiazolidinedione] for the primary reason of HCC prevention," said Dr. El-Serag.

"I don’t think the evidence is convincing at this point" regarding preventing HCC, said Dr. Caldwell. "The thiazolidinediones seem to retard progression of NASH fibrosis, but they also have adverse effects and their popularity has decreased."

Early days for a complex pathology

It seems as if the links between obesity, diabetes, and metabolic syndrome and NAFLD, NASH, and liver cancer are so tangled that it will take more time to fully resolve the etiologic relationships and the implications for diagnosis and management. The bottom line today is that a growing segment of American adults face risks for significant liver disease because of obesity, type 2 diabetes, and other elements of the metabolic syndrome.

"We see more and more patients over the last decade with liver cancer who didn’t have hepatitis or alcohol use but have diabetes and obesity. It’s a changing demographic," said Dr. Hezel. "We increasingly see liver cancer in patients without one of the classic risk factors. There are two possible mechanisms. Fibrosis and inflammation" caused by NAFLD and NASH trigger cancer formation and growth, "or it could be a more direct effect of high insulin levels or other hormonal effects. This is an emerging area; it follows on the epidemic of obesity and diabetes."

Dr. Cusi, Dr. Caldwell, Dr. Baffy, Dr. El-Serag, Dr. Busuttil, and Dr. Hezel all said that they had no relevant disclosures.

m.zoler@elsevier.com

On Twitter @mitchelzoler

Obesity, diabetes fuel liver disease epidemic