Vantage point – Huge hep C costs prompt search for when and where to treat
The limits placed on the new generation of hepatitis C drugs like Sovaldi and Olysio have pointed up a little-discussed topic in hepatology: not every patient will experience severe complications from chronic infections.
Much of the evidence suggests that only 10-25% of patients will develop cirrhosis over a 20 to 30 year period, leaving a significant share with preserved liver function. Whether to treat the remainder is an ethical and economic test, with the $84,000 cost of Sovaldi becoming a huge barrier to wiping out the virus altogether. This raises an important question of whether there is a subset of patients who may not need any intervention.
“We think so far that only one out of four patients experience progressive liver disease. There are still patients who for years have normal [liver enzyme levels] and no fibrosis progression,” said Michael Manns, chairman of the department of gastroenterology, hepatology and endocrinology at the Medical School Hannover in Germany. “There will be a debate because this will be a cost issue in the future. I’m sure that in many countries, although the therapies are approved, they will only reimburse for patients with significant fibrosis.”
Focussing on the sick
Payers around the world are grappling with these issues now as new hepatitis C antivirals, including Gilead Sciences’ follow-up to Sovaldi, prepare for launches in coming months. With drugs offering the promise of cures for more than 90% of patients, hepatologists could significantly reduce infection rates in a very short time, if cost were no issue.
Cost is clearly an issue – to use Sovaldi to treat the 3.2 million Americans infected with hep C would cost $269bn, more than the nation’s entire $263bn prescription drug bill in 2012. This is why payers and insurers for now have typically authorised Sovaldi and Johnson & Johnson and Medivir’s Olysio – a $50,000 drug – only for those patients who are showing evidence of progression toward advanced fibrosis, cirrhosis or need for liver transplant.
US hepatologists themselves are still dealing with how to use these new drugs. Within days the American Association for the Study of Liver Disease and the Infectious Diseases Society of America will release a new section of their hep C guideline titled “When and In Whom to Initiate Therapy.”
There are two data points that might shed some light on the value of these antivirals: the so-called “number needed to treat” analysis, or NNT, that is sometimes done with healthcare interventions to develop cost-effectiveness analyses; or study of predictive markers that could help identify patients more likely to progress.
The former of these, NNT, is not easily found in the scientific literature. These sorts of analyses typically state the number of patients who need to be treated in order to avert one complication – in the case of hepatitis C, such an analysis would be the number of patients taking an antiviral to avert a single case of cirrhosis, liver cancer or transplantation.
The UK’s National Institute for Health and Care Excellence (NICE), in drafting a Sovaldi guideline, has disclosed something akin to this. Drawing from the literature, the cost-effectiveness agency established that at age 30, a non-cirrhotic patient infected with hepatitis C genotype 1 has a 0.6% chance of developing early liver disease called compensated cirrhosis in the next year. This annual probability rises with age to 1.6% at age 50 – thus the chance of disease progression accumulates and rises as long as patients remain chronically infected.
“For many years [patients can] appear to be very slow or non-progressors,” said Charles Gore, chief executive of the Hepatitis C Trust in the UK. “Then typically in the 50s it starts to accelerate and it can accelerate very rapidly.”
And payers clearly know the cost of liver disease – in 2011, Michigan’s Henry Ford Health System calculated the cost of hep C patients with compensated cirrhosis at $22,752 a year, rising to $59,995 for those with end-stage cirrhosis. This explains why these patients have been prioritised for antiviral treatment.
But without an explicit NNT analysis in the healthier population, it is tough for payers to make a cost-benefit analysis. Part of the reason is the drugs are so new, as acknowledged by Express Scripts, the largest pharmacy benefit manager in the US.
“Express Scripts and our research partners are pursuing a number of studies regarding clinical outcomes and cost effectiveness of hepatitis C therapy in real-world situations,” Express Scripts spokesman David Whitrap said in a statement emailed to EP Vantage. “As Sovaldi just launched in December 2013, we still lack enough outcome data to share at this time.”
For its part, Gilead prepared its own cost-effectiveness estimates for NICE that found, for example, that a genotype 1 infected patient treated with Sovaldi plus interferon and ribavirin for 12 weeks would cost the National Health Service less than £15,000 ($25,248) per quality adjusted life-year, below the £20,000-30,000 threshold at which the agency usually will deny coverage.
The assumptions from that analysis are likely favourable to Gilead’s case, and NICE’s own estimates will almost certainly find that Sovaldi is more costly. In any case, in its draft guidelines NICE has stated that it is “minded not to recommend” treatment with Sovaldi because of its absolute cost – it asked for more detailed cost-effectiveness data from Gilead by disease progression status, genotype, treatment history and HIV co-infection status, with a target of publishing a final guideline by October 2014.
The second data that could help are predictive markers identifying which patients are more likely to progress, something that would be a holy grail to payers, said Prof William Irving, a virologist at the University of Nottingham.
Prof Irving, who has studied disease progression as part of the Trent Hepatitis C Cohort Study begun in 1992, said he had attempted to study non-progressors. The difficulty here is that to accurately measure progression, repeated biopsies are necessary – something that is difficult to ask trial participants.
Furthermore, the size of the biopsies, representing 1/150,000th of liver tissue, means that signs of developing fibrosis can be missed. Further confounding this analysis, he said, was the lack of a control group and an inability to adjust for fibrosis resulting from normal aging.
“Using a strict definition of non-progressors, there weren’t many,” he said of that evaluation.
However, the work of this regional longitudinal study – now merged into a national cohort – revealed one important non-biological indicator of patients who may not progress quickly, Prof Irving said. “The best single marker is how much alcohol they drink,” he said.
Patients within the cohort who attended clinic usually took the advice not to drink and did manage to limit progression, he said. However, a no-alcohol message is not a feasible intervention for the entire population. “We know as a whole, the (hep C)-infected patient group is quite a difficult one to engage with, and many patients don’t attend clinic appointments or lead a very chaotic lifestyle,” he said.
Prevention as the cure
A final question is how effectively a cure can prevent further spread of disease. As the University of Hannover’s Prof Manns puts it, “If you’re a dentist, you certainly want to get rid of the virus.”
In fact, a strong case can be made for targeting younger hep C patients who inject drugs because of the risk of enabling its spread through the sharing of needles. Combining the new direct-acting antivirals with opium substitution therapy and needle exchange can halve hep C prevalence, according to a model published by British researchers.
But it raises questions about how a healthcare system prioritises care with limited resources, a decision that for now payers have dealt with very differently. “If you treat a 60-year-old man with cirrhosis who’s got hep C it’s very unlikely he’s busy infecting anybody else,” said Prof Irving. “If you treat a 25-year-old drug user who’s still actively using then you treat that individual but you also potentially save a whole pile of new hepatitis C infections.
“The health economists just deal with the numbers and as I understand the analysis it is more cost efficient to treat younger, fitter people,” he said. “But clinicians treat patients, and it will always be a priority to treat the patient who you feel really needs it now.”
Should researchers be able to identify an additional sub-population of non-drug-using patients who will not accelerate toward cirrhosis as they age, hep C treatment could become more cost-efficient. Such a discovery would change the conversation between big pharma and payers.