Replies to post #166769 on Biotech Values

[ronpopeil]

Very nice analysis. I would be happy if enta hit that level

[biocqr]

ENTA has approximately 19.7M fully-diluted shares (including about 1.8M in the-money options). Dividing $1.3B by 19.7M shares gives a target price of $66/sh.

Is that your target today? If not what is your time frame. I didn't notice any DCF analysis in your model.

[oc631]

the maximum throughput of the HCV medical establishment, which is approximately 350K patients per year.

This is the most interesting part of your analysis. Where did you come up with the 350K patient figure? Is this a trailing figure based on interferon-based treatment?

[biotech_researcher]

ENTA Valuation Model- Nice work, however it begs a question...

You show (I believe) peak sales.. don't you have to discount back to today's value?

Example:

ENTA’s share of ABBV’s HCV royalties is tiered according to sales, with the lower tier paying about 4.5% (#msg-91918990). Thus, ENTA’s royalty on $3B of ABBV sales—at the lower royalty tier—is $135M per year. However, at the level of ABBV’s HCV sales I envision, some of ENTA’s ABT-450 royalties will likely be at the higher tier, which is an increase of about 50% relative to the lower tier (#msg-91918990). Since the sales thresholds for triggering the middle and high royalty tiers have not been disclosed, I’ll rather arbitrarily increase the $135M figure above to $150M to account for the multiple-tier effect.

[mcbio]

I estimate that ABBV’s 3-DAA regimen will garner about 100K of the available 260K patients per year in the US and EU, a market share of 38%. (Most of the rest will initially go to GILD since few patients will opt for interferon-based regimens.)

Respect your opinion, as always, but I don't see ABBV/ENTA's 3-DAA regimen (4 or even 5 drugs total) capturing a market share of 38%. I'll go on record for a much lower market share. I've got no prob eating crow here if I'm wrong. ; )

[ciotera]

ENTA Valuation Model

What makes you think that there is a treater capacity constraint? Do you have any supporting evidence for this? If treatments become so simple, couldn't the treater pool expand to PCPs who don't treat now?

But more importantly, what makes you think that the actual annual patient demand could ever get so high as to exceed this capacity?

And even more importantly, what makes you think that if this category grows to that level of annual expenditure, the payers will not limit treatment to F3/F4s like they do in Brazil or Canada already? I would, if I were them.. what's the point of spending so much money to treat minimal disease that only has single digit chance of leading to HCC? Why not wait and only treat those who have higher chance of progression since efficacy is only marginally worse in cirrhotics?

Overall, I think you are over-optimistic about this market, just like every analyst out there. For instance, there is no reason to believe that IFN-free regimens are going to change the fundamental boom-and-bust dynamic (unless there is a payer or infrastructure constraint). In fact, they are likely going to amplify it.

[turtlepower]

Thanks for posting this. Has ENTA ever disclosed whether the milestones are tied to any special circumstances such as first to approval or being part of a 2 DAA instead of a 3 DAA? Would they have to disclose it if they haven't already?

[kris_kade]

Thanks for posting the ENTA Valuation Model.

Any reason ENTA is still not up there with MNTA and ARRY on the "Read me first" links ? :-)

[DewDiligence]

Three verys: ABBV/ENTA’s HCV regimen has the potential to be “a very, very, very big” product, said Bill Chase (CFO) at ABBV’s CS webcast today. My own projection of peak sales is $3B (#msg-92234861), so I guess that qualifies, LOL.

[DewDiligence]

Re: HCV pricing and compulsory licenses

Cost per gram is a preposterous metric for evaluating drug pricing; nevertheless…

http://hepatitiscresearchandnewsupdates.blogspot.com/2013/11/will-new-hepatitis-c-drugs-trigger_26.html

In the analysis, [Andrew Hill of Liverpool University] and a colleague assumed that the production cost per gram of a hepatitis C drug was between one to 10 times higher than the equivalent HIV antiretroviral, depending on the complexity of chemical synthesis. The production cost of sofosbuvir, for instance, was estimated at $68 to $136 for a 400mg dose for 12 weeks. A Gilead spokeswoman declined to comment on the research and added that it is “premature” to comment on pricing.

This kind of argument may resurface when all-oral regimens are approved in emerging markets and governments have to decide whether to execute compulsory licenses as permitted under WTO regulations. India and Brazil would seem to be likely candidates for such action.

Note: My valuation model of ENTA (#msg-92234861, #msg-92239448) ascribes $100M of value to ABT-450 royalties outside the US and EU, but almost all of this is attributable to Japan, leaving very little value for royalties from emerging markets. This is due in part to the threat of compulsory licenses.

[DewDiligence]

ENTA ($29.66, +7%) new all-time high (eom).

[DewDiligence]

ENTA Valuation Model

[Revised pricing assumptions as a consequence of Sovaldi’s US list price of $84K.]


Let’s first simplify the analysis by considering only the HCV market in the US and EU. During the next several years, the market for HCV drugs in the US and EU will be limited not so much by the number of patients with chronic HCV infection, but rather by the number of treating physicians and how many prescriptions they can write per year—i.e. the maximum throughput of the HCV medical establishment, which is approximately 350K patients per year. Of these, about 260K patents have genotype 1a or 1b, which is the patient pool ABBV/ENTA are addressing in their initial NDA/MAA.

I estimate that ABBV/ENTA’s 3-DAA regimen will garner about 100K of the available 260K patients per year in the US and EU, a market share of 38%. (Initially, almost all of the remaining 62% will go to GILD’s Sovaldi + Ledipasvir combination because few patients will opt for interferon-based regimens.)

I conservatively model an average selling price (net of discounts and rebates) of $41K for a 12-week course of treatment with ABBV/ENTA’s 3-DAA regimen; this $41K figure assumes an ASP of $25K in the EU and $60K in the US, with 55% of the US+EU patients coming from the EU and 45% from the US. (Sovaldi’s US list price is $84K for 12 weeks of treatment; GILD has not disclosed expected pricing for Sovaldi in the EU, nor has it publicly discussed US pricing for the Sovaldi + Ledipasvir combination.) 100K patients per year treated for 12 weeks at an average price of $41K equates to $4.1B of annual HCV sales booked by ABBV in the US and EU at peak throughput.

Some patients taking ABBV/ENTA’s 3-DAA regimen, such as those with cirrhosis, may require 24 weeks of treatment rather than 12 weeks; however, to be conservative, I’m ignoring the incremental revenue from 24-week treatment durations and hence I make no adjustment to the above calculation for treatment duration longer than 12 weeks.

ENTA’s royalty on ABBV’s sales of ABT-450 is tiered according to product sales in the same manner as IRS tax brackets; according to ENTA, the lowest-tier royalty rate is in the low teens, and the highest-tier royalty rate is approximately 20%.

Because ABBV will charge an integrated price for the 3-DAA cocktail of ABT-450, ABT-267, and ABT-333, ABBV’s sales of ABT-450 per se have to be calculated according to a contractual formula that is based on the proportion of “value” ABT-450 contributes to the overall HCV regimen, which I estimate is 35%. Why 35% rather than 33.3%? Because ABT-450 (the protease inhibitor) and ABT-267 (the NS5A inhibitor) are obligate components of the cocktail, while ABT-333 (the non-nucleoside polymerase inhibitor) can be viewed as “supplemental insurance” against resistant mutations. (Indeed, ABBV/ENTA are testing the 2-DAA combination of ABT-450 + ABT-267 in a phase-3 trial in Japan (#msg-91870291) and in various phase-2 studies.) It matters little whether ribavirin ends up being included in ABBV/ENTA’s 3-DAA regimen for some or all genotype-1 patients because ribavirin is available in all countries as a cheap generic.

Combining the 35% figure in the above paragraph and the information about royalty tiers in the preceding paragraph, the net result is that ENTA’s average royalty rate on ABBV’s sales of its 3-DAA regimen is about 5%. Applying this 5% rate to ABBV’s $4.1B of peak annual sales of the 3-DAA regimen yields $205M of peak annual royalty income for ENTA.

Since ENTA is not responsible for any development or commercialization costs related to ABT-450, the $205M annual royalty calculated above is also ENTA’s annual pre-tax net income attributable to ABT-450 in the US and EU.

To be conservative, I’ll assume that ENTA pays the full statutory tax rate on all income even though we know that some income will be offset by tax-loss carryforwards. After paying the statutory tax rate, ENTA’s $205M annual haul is reduced to about $127M.

Although there’s a bolus of HCV patients who can be expected to seek treatment with an all-oral regimen as soon as one is approved, I don’t expect the all-oral HCV market to exhibit a parabolic boom and bust the way VRTX’s Incvek did because the number of patients treated per year is constrained by the throughput bottleneck described at the beginning of this analysis. Thus, it’s likely that the all-oral HCV market will peak relatively soon after launch and then begin a gradual decline with a robust market continuing until well into the 2020’s decade.

Given the kind of sales trajectory described above, I apply a multiple of 8x to ENTA’s $127M peak annual after-tax royalty income. This implies a value of $1.02B for the series of royalty payments to be received by ENTA for ABT-450 in the US and EU.

What about the rest of the world—i.e. geographies other than the US and EU?

I conservatively add $100M of value for ENTA’s ABT-450 asset in Japan and all other countries aside from the US and EU. The $100M figure is conservative because: i) Japan is a large and affluent country; and ii) ENTA’s royalty rate is higher in Japan than in the US and EU because ABT-450 comprises a higher proportion of the overall regimen’s value when there are 2 DAA’s in the regimen rather than 3 (#msg-91870291).

Getting back to the US and EU… if ABBV/ENTA’s 3-DAA regimen is merely approved for marketing in these jurisdictions, ENTA will earn almost $195M in milestone payments from ABBV (see #msg-91918990), so let’s add that in.

What about ENTA’s HCV partnership with NVS?

I add $100M for the value of ENTA’s EDP-239 partnership with Novartis, which has recently shown some life (#msg-89752221). I think $100M is a conservative figure in light of the valuations of such competing HCV companies as ACHN and IDIX, especially when one considers the lofty milestone payments ENTA will receive from NVS if the program pans out (#msg-89280314).

I add $50M for the value of ENTA’s second-generation HCV program with ABBV (#msg-89280030) and $25M for ENTA’s antibiotic program partnered with NIAID (#msg-91629594).

All told, the above components of value come to $1.49B for ENTA’s non-monetary assets. Adding in the $112M of cash on the balance sheet as of 9/30/13 (#msg-94395530) gives $1.60B as the bottom-line figure. (As noted above, I’m conservatively ignoring the economic value of ENTA’s tax-loss carryforwards.)

ENTA has approximately 19.7M fully-diluted shares, including about 1.8M in the-money options. Dividing $1.58B by 19.7M shares yields $81 per share.

Please note that this $81/sh figure is not discounted for the probability of program failure or the value of time. I’ve deliberately not applied such discounts because each investor is apt to have his own ideas of how such discounting should be done. Thus, the $81 figure represents my assessment of the fair value of ENTA at the time of product launch of ABBV’s 3-DAA regimen (expected to occur in early 2015), assuming that there has been no significant progress on any of ENTA’s other programs (including the partnership with NVS) that would warrant recalibration of the model.