DrugBaron has, in the past, been in a minority of voices putting forward the arguments in favor of these high prices, in the face of a wall of criticism from patients, the wider public, and senior government figures such as Senator Waxman, who waded into the Solvadi pricing row by writing to the chief executive at Gilead asking for a re-think on pricing couched in a mixture of pleading and threats.

There are two prongs to the argument in favor of high pricing: one pragmatic and one philosophical. From a practical perspective, high prices are necessary to justify the investment required to develop new medicines. Particularly in rare diseases, if prices were lower, no one would invest in innovation and patients would suffer. While DrugBaron has also argued that systemic inefficiencies in the drug discovery process are the root cause of these high prices, until such structural inefficiencies are addressed there will remain a stark choice between paying the high prices or leaving individuals with rare, but often serious, conditions essentially untreated.

Equally importantly, there is a moral obligation to protect the rights of the inventors of such innovative therapies to charge whatever they deem appropriate. To do otherwise threatens to undermine the entire patent system, whose central bargain is to make an innovation publicly available in return for a short period of monopoly. Patients and healthcare providers have the absolute right not to buy the product at the stated price, but should not have the right to demand its provision at a lower price.

Indeed, for many medicines that have commanded a high price in recent times, the degree of efficacy has been such that it is unclear whether the price, though numerically large, is really excessive. A crude comparison between the cost of an early prophylaxis against smallpox in the 1750s and Solvadi in 2014 suggested the perceived value of a curative against an otherwise fatal disease has remained remarkably constant across centuries and continents.

But in this regard, Entresto is also an outlier. It is nothing even resembling a cure.

A quick review of the clinical trial data is pretty straightforward: Entresto was approved on the basis of a single, large Phase 3 trial called PARADIGM-HF, in which 8,442 patients were randomized to the Entresto combi-pill, or to an active “standard of care” arm consisting of the ACE inhibitor enalapril. The trial was stopped early by the data monitoring committee, and the evidence of superiority was unequivocal. The primary end-point (a combination of cardiovascular death or first hospitalization for heart failure after treatment commenced) was significantly less frequent in the Entresto group (the hazard ratio was 0.80, meaning the chance of harm was 20% lower in the Entresto group, with a p value less than 0.001).

That Entresto is superior to 10mg of enalapril twice daily in this patient population is beyond doubt. But what does it really mean in the real world?

Firstly, while ACE inhibitors are the recommended first-line therapy for the majority of individuals in this patient population according to the 2013 AHA treatment guidelines, a significant minority would have received alternative treatment combinations outside the clinical trial setting. Moreover, even among those treated with enalapril, some at least would receive 20mg twice daily as opposed to the 10mg dose used in the PARADIGM-HF comparator group. In short, the comparator group in PARADIGM-HF, being a “one-size fits all,” likely did worse than real-world standard of care.

From a simple, clinical perspective one could argue that doesn’t matter much: a “one size fits all” regimen with Entresto was better, so why not use that? But from a pricing perspective it matters a great deal. If the Entresto regimen (at $4,500 per annum) was actually only the same as can be achieved with a tailored selection of generic medicines at a tenth of the cost, then the price premium is clearly not justified.

Secondly, Entresto is a combination of two agents: one, valsartan, was itself a mega-blockbuster for Novartis, racking up annual sales in excess of $6 billion as Diovan prior to its patent expiry in 2012, and is now available as a generic at less than a dollar a day. The other is a new molecular entity, the neprilysin inhibitor sacubitril, a first-in-class agent not previously approved for any indication. Since angiotensin receptor blockers (ARBs) like valsartan have been used as first-line treatment for heart failure, and have shown significant benefit in clinical trials, it’s reasonable to wonder how much of the benefit of the combination is due to the valsartan. If most or all of the benefit could be achieved with dollar-a-day valsartan, once again the price premium for Entresto would be hard to justify.

Unfortunately, though, because PARADIGM-HF was performed with an ACE inhibitor as the comparator, we don’t have a definitive answer to this question.

In the FDA briefing documents, the clinical reviewers also speculated around this question. Although they concluded that a contribution towards the efficacy from the sacubitril component was likely, the magnitude of the contribution remains impossible to quantify without a direct comparison. With the approval of Entresto, however, it is unclear whether such a study will ever be performed.

Once again, though, clinicians and payers alike are left without the necessary information to estimate the true cost-effectiveness of Entresto.

Thirdly, there is the question of the absolute benefit to the individual patient. A 20% reduction in cardiovascular mortality during the median follow-up of 27 months in PARADIGM-HF sounds impressive. But those deaths are delayed a little rather than prevented. A reliable extrapolation of the Kaplan-Meier curves suggests at most a 6-month median benefit (which, as noted above, is already likely to be an over-estimate of the true benefit of Entresto versus an optimal standard of care using generic drugs).

Given that NYHA stage II-IV heart failure with ejection fraction below 40% (the criteria for enrollment into PARADIGM-HF) has a median survival less than 5 years, it bears comparison with many cancers in terms of the mortality burden. High-priced cancer medicines that delay progression by a few months, but do not represent a cure, have also come in for considerable criticism.

Unlike Soliris or Sovaldi (and perhaps even Repatha and Praluent, although hard evidence of their real efficacy on cardiovascular events has yet to be collected), Entresto is no miracle cure. On the other hand, it certainly does have efficacy – PARADIGM-HF was carefully designed to have the maximum chance of demonstrating that, and on that basis the FDA was surely right to approve it. Between those two extremes, though, it all comes down to price.

Unfortunately (and one suspects it is no accident), PARADIGM-HF provides insufficient evidence to reliably estimate the true value of Entresto, making a quantitative assessment of its cost-effectiveness almost impossible.

Interestingly, the Boston-based Institute for Clinical and Economic Review (ICER), a non-profit research organization, concluded last week that the list price of Entresto fairly reflected its clinical benefit, and could be considered “cost effective” (although they did note that with at least two million eligible patients, this would create a budget impact so high as to place excessive cost-burden on the healthcare system as a whole, and lead to material growth in health costs compared to national income). Importantly, though, ICER’s analysis modeled the benefit of Entresto using the PARADIGM-HF results – effectively ignoring the possibility that much of the benefit was illusory, due to the trial design.

For its part, Novartis has said it will consider performance-based pricing, with discounts for patients where the desired outcome is not achieved and a bonus where it out-performs expectation. But the details are still sketchy, and it’s entirely unclear how you would assess the performance of the drug versus standard of care on an individual-by-individual basis, particularly when the expected benefit is just a couple of extra months prior to hospitalization or death.

But without new data (none of which is on the horizon), it may be very difficult for clinicians and payers to resist the pressure to provide legions of heart failure patients with Entresto. A step-change in the treatment of heart failure it is not, but a blockbuster it will likely emerge nonetheless. The billion-dollar question (quite literally) for payers is how to stop clever clinical trial design by sponsors translating into vast unearned premiums being paid for unproven superiority in the future.