According to published research the average POS of Ph III orphan drugs is between 47% to 66% (refer my earlier post). So you're right that 30% seems low. Although if you just looked at Orphan CNS historical averages you might say its about right (refer same post).
However, I think royalwulff55 is right that the strength of the Phelan-McDermid results suggests the POS should be higher. Jon has emphasized that this Ph II result well exceeded the threshold for success. He stated that future trials did not need to achieve this level to still be considered successful. The fact that NNZ-2591 went well over the bar should increase it's Ph III POS (as well as the Ph II likelihood of success for the remaining trials).
An additional positive factor not on royalwuff55's list is that all the diseases currently being targeted can, I believe, be diagnosed with 100% certainty using genetic testing. This is not the case with many diseases (rare or otherwise). Being able to identify your patients accurately should increase the likelihood of the drug showing success in a high percentage of patients. If the disease being treated is diagnosed using just symptomatic or blood test criteria you can't be 100% sure you're treating the same underlying condition in all cases. A good example is cancer - which has, as I understand it, turned out to not be one thing, but hundreds of different 'sub-types', each of which responds differently to treatments. Some of the greatest recent successes in cancer treatment have come, as I understand it, from targeting specific genetic mutations rather than just 'lung cancer' or 'bowel cancer'.
Putting all this together it seems reasonable to suggest the failure risk of Phelan McDermid should be less than half that of the 'average' Orphan drug.
47% to 66% represents a 53% - 34% risk of failure. So halving the risk of failure would increase our POS to between 74% to 83%.
By this train of logic a 75% POS for Phelan-McDermid might be a reasonable value. royalwuff55's large list of positives might argue for higher. Although I thought Hashan's report also made some good arguments for why the Ph 2 trial should still be taken with a dose of salt. Objectively the Ph II results cannot be taken as predicting a guaranteed likelihood of Ph III success, for all the reasons he listed.
However, from a valuation perspective a 75% POS is already so high that it would make little difference increasing it further. For example, if you say Phelan McDermid has just not a half but a quarter the likelihood of failure of the average orphan drug you might argue for a POS of 87% to 92%. Burt in any valuation model this only represents further increase in value of 10% - 17%. In other words, from a valuation perspective NNZ-2591 for Phelan-McDermid is already pretty much there.
Put another way, when it comes to valuing NNZ-2591 for Phelan-McDermid the uncertainty left in the POS (10-20% on above logic) is dwarfed by the much wider uncertainties that still remain regarding drug pricing, number of patients and market penetration.
The above logic suggests that if NNZ-2591 passes Ph 2 in any Orphan indication with good results then the drug should be valued pretty much as if approval is likely.
One thing that bolsters the above logic significantly is that Neuren is not running a '1 shot' strategy, but currently has 4 'shots' at success. If we look at how this changes things (from a modelling & valuation perspective), we get the below probabilities of approval (rounded to nearest 1%) using the above scenarios.
| Number of indications approved | Using 50% POS from Ph III | Using 75% POS from Ph III |
|---|
| 1 | 0 | 6% | 0.4% |
|---|
| 2 | At least 1 | 94% | 99.6% |
|---|
| 3 | At least 2 | 69% | 95% |
|---|
| 4 | At least 3 | 31% | 74% |
|---|
| 5 | All 4 | 6% | 32% |
|---|
So if the Ph II's are all successful you can take it as almost a given at least 1 indication will reach market.
This of course assumes the trials are all independent, which they are not. Because they're not independent the success or failure of one indication significantly increases or decreases the likelihood of success of the others. For example, its unarguable the success of the Phelan-McDermid Ph II trial increased the likelihood of success of the remaining 3 Ph II trials (although exactly by how much is hard to judge accurately).
However, for modelling purposes its easiest (and reasonable) to consider them all as independent for now. In which case the likelihood of a certain number of successful indications is as follows:
| Number of indications approved | Using 50% POS from Ph III | Using 75% POS from Ph III |
|---|
| 1 | 0 | 6% | 0.4% |
|---|
| 2 | Exactly 1 | 25% | 4.7% |
|---|
| 3 | Exactly 2 | 38% | 21% |
|---|
| 4 | Exactly 3 | 25% | 42% |
|---|
| 5 | All 4 | 6% | 32% |
|---|
In conclusion, if all 4 Phase IIs are successful, then the likelihood of Neuren then failing to bring any of them to market seems very small. At that point Neuren should therefore be worth a minimum of $130 to $210 per share (using the earlier analysis I posted using 30% market penetration + 50% for ROW + 50% for additional indications).
(20min delay)