Geron: Of Abstracts And Abstractions

About: Geron Corporation (GERN)
by: BioTrench


Abstracts for myelofibrosis (MF) and myelodysplastic syndromes (MDS) are positive and suggest efficacy.

MF median OS has sort of been reached, depending on how data is analyzed.

MDS data is positive, but a competitor abstract leads us to believe that GERN will continue to pursue a sicker population.

Possibly long execution timeline, dilution risk and terse management poses barriers.

With the recent release of the ASH abstracts and GERN's conference call, a flicker of light has been shed on the situation. GERN has managed to obtain two oral presentations for their trials in myelofibrosis (MF) and myelodysplastic syndromes (MDS), indicating that the reviewers believe the results to be significant. While GERN's abstracts contain information that can be considered common knowledge for close followers, there are a few points of interest which we will discuss herein. More importantly, the overall thesis that imetelstat is effective in both MDS and MF is intact, as shown by a market rally of 20% on the day the abstracts were released. Analysis will chiefly revolve around the MF and MDS abstracts along with the recent conference call. Note that the data presented in the abstract was from Q2 2018. During the conference call, it was stated that updated details will be provided during the ASH presentation. While there is more waiting ahead, there are a few important things to pay attention to at this point in time.

MF MOS has sort of been reached

Median overall survival (MOS) has become the focus of the MF trials, but it's a bit of a hit and miss. Two dosing regiments were used for the trial - 4.7 and 9.4 mg/kg. It's been known that the 4.7 mg/kg arm was largely ineffective, but now we know that the MOS was 19.9 months, which is slightly better than literature values for MOS in the JAK-inhibitor relapsed/refractory (r/r) population (~14 months). More importantly, the 9.4 mg/kg arm is still going strong. However, there is an important, and nuanced, point to note.

Roughly 25% of the study population was considered to be triple negative (not containing a mutation in JAK2, CALR, MPL genes). Non-triple negative patients had a MOS of 23.6 months (75% of study population), whereas MOS had not been reached in the triple-negative population. This would mean that at least 37.5% of patients passed away by the 24 months mark. Assuming for some deaths in the triple-negative population, this number is likely to be in excess of 40% by 24 months (around May 2018).

Per the conference call, we would then likely see MOS results by the time ASH rolls around. This would mean that there's a good likelihood that overall MOS is 30 months or less, predominantly guided by the non-triple negative population. The good news is that it's effective in a specific subset of patients. The bad news is it's a tiny number population wise. MF prevalence has been estimated to be between 2-4 per 100,000 people, which puts the US MF population at ~10,000 patients in a given year. Triple-negative accounts for roughly 9% of the MF population, which would be approximately equal to 1000 patients. With such sparse numbers, it may be difficult to recruit patients for a trial targeting this population, if and when the time comes. Perhaps this aspect is a tripping point for how to progress forward, as they continue to follow-up patients and watch the situation unfold.

MDS results positive as expected but competition may impose limits

As with our previous assessment of the MDS trials, overall, everything is quite positive; however, competition casts a slight cloud over the results. By and large, the trial can be summarized in three points: 1) Patient population relapsed/refractory to erythropoiesis-stimulating agents have few options, 2) high transfusion burden (8 units/8 week) in study cohort means they're quite sick, and 3) 8 week+ transfusion-independence in this population is better than anything else in the literature, even with immature data.

Instead of imetelstat in MDS, it is more interesting to look at an abstract for luspatercept which is targeting the same population. Their phase 3 trial had a transfusion baseline of 5 units/8 weeks, and an 8 week+ transfusion independence rate of 38% (versus 8% in placebo). The good news is that imetelstat managed to have a similar performance in a sicker population. The bad news (for GERN) is that luspatercept's toxicity profile appears to be better. With an earlier shot at approval, it will likely end up being the preferred first choice for less sickly r/r patients before moving on to imetelstat. Whether this was a contributing factor as to why GERN went for a sicker population, or whether it was just a stroke of luck, is unknown; however, it is a factor to keep in mind going forward.

A 5 year game plan?

GERN reconfirmed their focus on MDS and cash sufficiency for at least a year or two, but the bigger question is how will execution pan out. Their current goal is to get MDS done within the next 5 years, with trial recruitment beginning in mid 2019.

Assuming that enrollment takes a year to complete, and the study runs for another two years for sufficient follow-up, we estimate top-line results from the Phase 3 portion of IMerge, to be available around mid-2022. From there, we believe preparation and submission of both new drug application or MAA within the European Union would need approximately six to nine months.

After that it would take approximately 12 months to get an FDA approval and 24 months to get an EU approval. With these assumptions, the estimated timeline for imetelstat’s first potential approval is approximately the end of 2023.

From: 2018 Q3 Conference Call

Overall, our expectations for this trial is largely positive, and, barring any unexpected events, should play out favorably. More importantly, we believe that 5 years is a conservative estimate. They still have a fast track for MDS, which could possibly play out in their favor.

Regardless of their speed, a possible dilution will be hovering over GERN, unless they get bought out or partner up. GERN has estimated that the MDS trial will cost $50 million. This will effectively reduce their current cash and equivalents (~$185M) to $135M. Currently, they are on track for burning $37M in 2018. Following a hiring spree, their future burn rates may be closer to $50M per year. This would leave them with enough cash to operate for ~2 years. Unless they partner up or some other major event happens, they will likely need to raise funds. Aside from funding, GERN has been relatively terse about other executional aspects. While there has been continuous mention of items such as transferring of data, hiring, and partnerships, little detail has been provided. This introduces a layer of ambiguity that makes it difficult to assess additional risks.

Concluding remarks

With the release of the ASH abstracts, we conclude that imetelstat is effective in both MF and MDS, but there may be a few obstacles in the near future. Depending on how GERN decides to pursue MF, they may be working with a really small population. Meanwhile, in MDS, competition may force GERN to continue running trials on relatively sicker patients. In the meantime, GERN has proposed a 5 year plan for MDS, but little is known about what management is planning to do with respects to everything else (operational and clinical). While we continue to expect ASH presentations to be positive, GERN's future post-ASH looks to be rather haphazard with a long investment time frame, which may pose undue risk.

Disclosure: I am/we are long GERN.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.