With the approval of Andexxa (Andexanet Alfa) came many headaches for those of us in the pharmacy world. The approval process was a bit odd and left us with many questions about safety, efficacy, cost, and post-administration anticoagulation. On top of that, it was made available to only select Level 1 Trauma Centers which needed to go through approval to bring it to formulary. Because it was approved so quickly, the FDA requires that PTLA continue to study the medication after marketing (ANNEXA-4 Phase 3b/4 study), and nobody really had any real standard processes to go off of, but providers seemed excited to have the opportunity to order the medication for reversal of anticoagulation with Apixaban/Edoxaban/Rivaroxaban.
Needless to say providers are very gung-ho about ordering this medication wherever possible. With increasing popularity of Factor Xa inhibitors and bleeding rates cited as frequently as 2.1-3.5%, there is a definite need for the specific reversal agent. There are also restrictions to its use at most hospitals due to cost ($37,500 to $50,000 - before upcharge to patients, a patient could actually be charged $200,000 for this medication when we do not have a ton of data on it), it takes a while to reconstitute using multiple vials, is stored in the pharmacy, and is used in urgent situations. So, when I say it's a headache for pharmacy departments - it really is. It takes a while to reconstitute and bring down to the floor/emergency department which does not mix well with emergent situations. But it has real utility in acute care settings and providers are excited to use it. On top of that, Portola Pharmaceuticals (PTLA) is expanding its availability to smaller hospitals, meaning more earnings potential and revenue growth. Portola has also received European approval for it under another brand name - Ondexxya. They also have the already approved Bevyxxa and investigational drug Cerdulatinib. Below, we'll dive into some recent news and financial reports, trends/reactions in share price and where Portola may be headed.
As of the time I am composing this article, the market has taken a significant downward turn with a slight correction upward on 6/5/19. There is no telling whether this will be a short-term correction with significant upside going forward, or an extended period of recession. As the 3-month/10-year yield has recently inverted, we probably have 1-2 years before a recession sets in, if we use 2006-2008 as a benchmark. That said, the market may correct and go on a bull-run over the next couple of years, especially if tariffs are removed. But, considering the President has just proposed new tariffs on to Mexico and extended the China tariffs by another two weeks, it is probably safe to assume we will be at least trading sideways. All that being said, I would still consider PTLA to be a speculative stock. Yes, they have FDA approved products, but they are not yet profitable. Portola seems to be on their way to profitability but should be hit hard by any lack of confidence in the market, especially considering the company has a beta of 3.88. For example:
During the October-December 2018 correction, PTLA followed the market very closely and ended at a low of $15.46 before rallying into April 2019 (Yahoo Finance):
Since the beginning of 2019, PTLA has rallied about 36% and currently stands at a value of $27.21. The stock peaked during this time in the middle of April-May at $37.67 but, since March and heightened trade tensions, has been beaten down:
It can be seen that PTLA tends to follow market/trader sentiment and may pull back more than other equities during corrections. Even after posting decent first quarter 2019 results, the stock pulled back in early May 2019 as can be seen on the chart above.
On May 8th, 2019, Portola announced first quarter 2019 revenue of $22 million for the period ending 3/31/2019. This was compared to $6 million for the same period one year prior and made up of $20.3 million in sales of Andexxa, $77 thousand in sales from Bevyxxa, and $1.8 million in licensure and collaboration revenues. Q4 2018 revenues from Andexxa totaled $14 million and $35 thousand for Bevyxxa. These results represent a 45% and 120% increase, respectively. It is reasonable to expect similar growth for Andexxa revenue going forward as Portola increases availability from only large level one-trauma centers, to other emergency departments world-wide. At present, after one year, the CAGR for Andexxa revenue growth is 42%. Continuing forward, this should provide a steady revenue stream for the company.
Somewhat interesting is that the company has backed off its marketing and promotion of Bevyxxa (Betrixaban) from prior years. The medication has utility, but is not used often as there are many cheaper options available inpatient. Betrixaban has similar mechanism of action to Apixaban or Rivaroxaban, being a direct factor Xa inhibitor, but it is specifically indicated for use inpatient for VTE prevention during extended stays. During the past year, I have seen this medication recommended a single time - for a patient who was developing fatty lumps, irritation, and rash around the injection sites from Enoxaparin and Heparin that had been used throughout the two-month stay to prevent VTE. Currently, I see this as being, probably, one of the only utilities for the medication given its cost (other than a patient absolutely refusing injections, which is unlikely during an extended stay, or label/indication expansion). In other news releases, Portola has expressed their priority is on advancing and marketing Andexxa with Bevyxxa taking a lesser priority. This is encouraging, as the company is clearly recognizing which therapy has the potential to be a blockbuster. The company has, instead, created 10 centers of excellence to carry Bevyxxa and establish models for other institutions to base their practice off of.
Net loss for the quarter was $78.2 million for Q1 2019, and $84.2 million for the same period prior. Part of the loss for Q1 2019 includes a $5.8 million charge for manufacturing of second generation Andexxa, and $3.9 million generation one supply charge as hospitals transition to the next generation product.
As of 3/31/2019 PTLA reports cash, cash equivalents, and investments at a total value of $322.8 million. The company also reports opening a loan agreement for a total of $125 million with an additional amount of $62.5 million received. The company forecasts this cash pile will be able to last through the end of 2020. While a loan agreement puts a damper on speculative success, it also affords the company ability to avoid diluting shareholders; something which can lead to lower valuations and largely negative sentiment toward a company.
Operating expenses for the period ending Q1 2019 totaled $95.8 million, up almost $4 million from the year prior. Portola reported that this increase was due to rolling out product, European Ondexxya approval, and "field force" related to their medications. The company reports that cost of sales was $7.2 million, Research and Development costs were $35.8 million (down from $60.1 million the prior year), and selling, general and administrative expenses were $53 million. I would be expecting Research and Development to hold relatively steady in the future as the company advanced Cerdulatinib and its selective Syk inhibitor through trials but may decrease as Phase 3b/4 research of Andexxa is wrapping up. The company currently has 67 million shares, resulting in GAAP loss of $1.17/share, non-GAAP loss was $1.02, and analysis of the difference between the two can be found in the earnings release.
It seems that, at current rates, PTLA should have enough cash to continue through the next four quarters of operation. Add in their prospective growth of revenue from Andexxa, and their forecast of cash flow through end of 2020 seems reasonable. The company does have debt, but also has marketable products that should grow in sales in the near future, especially with current statistics on oral anticoagulant use. Portola also received a C-code reimbursement from CMS for Andexxa. While this does provide an avenue for reimbursement for Andexxa, it is only a temporary pricing code lasting somewhere in 3-5 years. The European approval of Ondexxya is also encouraging. This should allow for more revenue growth over time, but in the short term, there may be an increase in costs.
Recent Trial Results
The ANNEXA-4 trial is not set to be completed until 2022, so it has yet to be updated with results on clinicaltrials.gov. In their press release, Portola says that 77/98 patients treated and evaluated for spontaneous intracranial hemorrhage (ICH) achieved excellent or good hemostasis after treatment with Andexxa. Importantly, the results were determined by an independent adjudicator, which means investigators, and the company should have no involvement with decisions for efficacy. In a different publication of results cited in the current one, Andexxa was reported to have 82% efficacy across all bleed types, not just spontaneous ICH. But the part that causes headaches across the board and issues for the FDA and CMS for full approval of reimbursement pathways is that,
"[w]ithin 30 days of enrollment, thrombotic events occurred in 14 patients with spontaneous intracranial hemorrhage (10.9%) and death occurred in 24 patients (18.8%), consistent with the full ANNEXA-4 study results and the high background thrombotic risk of the enrolled patient population."
However, PTLA also reported that among the 18 patients who restarted oral anticoagulation within 30 days, there were no thrombotic events reported. The company is correct with the assumption that the patients do have higher background risk of thrombosis/emboli. However, it is the cost of the drug and no clear and universally accepted guidelines to re-initiating anticoagulation after treatment with this specific product. So, the argument that is often raised is: Yes, even though this is a high risk population for clotting in general, and these ICH patients will often re-present with a cerebral infarction after their initial ICH (about 2% will have recurrent ICH and 2% new ischemic stroke), almost 20% of patients will not survive in the next month. Considering this with the exorbitant price, it should be obvious why anyone would be reluctant to pay for this treatment.
However, as cited in the article above, ICH has a high early-case mortality, and it seems that Andexxa is able to achieve hemostasis in a large percentage of patients to prevent expansion. Mortality varies greatly with size and location in ICH. As many as 50% of deaths may occur in the first two days, and 51-65% of patients may die in the first year of followup after an ICH. Considering this data, it seems that Andexxa is reducing the death rate by over 50%, but we do not have extended followup results yet. For many patients, the most important thing is that only 20% of patients are expected to be able to live independently after an ICH. If treatment with Andexxa is able to provide more quality life years, then it may warrant the high cost (Not to mention the extensive research/development/costs it took to make the molecule).
Bevyxxa, on the other hand, is very similar to Rivaroxaban/Apixaban/Edoxaban in its mechanism of action, but with a different indication only for prevention of VTE with extended stay for acutely medically ill patients. The studies are somewhat controversial to me when I look at them. The first I looked at called EXPERT studied Betrixaban 15mg and 40mg orally twice daily for 10-14 days vs Enoxaparin 30mg subcutaneously twice daily in patients who received knee arthroplasty. Rates of VTE between the groups were 20%, 15.4%, and 10% respectively - it is important to note the approved dosing of Bevyxxa is currently 160mg by mouth on day 1 and 80mg once daily for 35-42 days. The trial results do not report any statistical significance here, this was a phase 2 study. So, data for this trial could be misconstrued as Bevyxxa not being beneficial, which it may not be at this dose. However, it is also important to note that the study reported rates of bleeding events for patients as with 0%, 2.4%, and 7.0% across groups. One may also, incorrectly, derive from this study that Bevyxxa will lead to lower rates of bleeding, but we cannot conclude safety differences from this study. This study is basically full dose enoxaparin versus less than full dose Bevyxxa, which would be expected to show less side effects.
In the EXPLORE-Xa study, Bevyxxa was tested in a phase 2 study to evaluate safety, tolerability, and (kind of) efficacy vs. Warfarin. Warfarin was dosed as we do normally in clinical situations - titrated in a patient specific manner. Bevyxxa was tested against it using 40mg, 60mg, and 80mg once daily - both arms were for a minimum of 3 months use. Groups seemed homogenous at baseline and were organized in a 1:1:1:1 manner. The most important comparison that should be considered is the 80mg Bevyxxa vs. Warfarin, because we now know that is the approved dosing. In clinically relevant, non-major bleeding, there was no statistically significant difference between Warfarin and Bevyxxa. For the secondary outcome of exposure adjusted incidence rate of any bleeding, the medication demonstrated a statistically significant reduction in any bleeding event. So, Bevyxxa is most likely not any more dangerous, bleeding wise, than Warfarin. The adverse event profile is comparable, and it undoubtedly has an anticoagulant effect.
The Phase 3 APEX study of Betrixaban 80mg daily for 35-42 days versus Enoxaparin 40mg subcutaneously for 10 +/- 4 days looked at acutely medically ill patients. Baseline patient data can be found in the NEJM publication, and across both groups, the BMI was approximately 29, and creatinine clearance >15-30ml/min for about 95% of patients. This information is important to me because it validates the study's use of fixed dose of enoxaparin - a medication that is commonly dose-adjusted for patients who are obese/underweight and those who had kidney dysfunction. An under-dosed enoxaparin regimen could inflate perceived efficacy of Bevyxxa. Primary outcomes included 3 separate cohorts consisting of all patients with D-dimer (a bio-marker that is not super specific for clotting events but can be a positive indicator) double the upper limit of normal, another with patients over 75 and the same biomarker levels, and all patients who received at least one dose of study drug without necessarily an elevated D-dimer. The fourth primary outcome was percentage of patients experiencing major bleeding through 7 days after discontinuation of drug. There was a non-statistically significant reduction in major bleeding, and there was statistically significant reduction in the composite primary outcome of symptomatic DVT, non-fatal PE, VTE-related death, or asymptomatic proximal DVT, through Visit 3 in cohorts two and three (actual implications are muddied, as cohort 1 was non-significant, the other 2 cohorts are considered exploratory, there is also a post-hoc study for critically ill patients showing a statistically significant benefit).
This study could also be confusing because disease states included were Congestive Heart Failure, acute infection without septic shock, acute respiratory failure, acute rheumatic disorders, or acute ischemic stroke with resulting paralysis. Some of these patients would be expected to be at higher risk for an embolic event, and considering Bevyxxa was administered for up to 42 days, and enoxaparin as little as 6, but up to 14, would suggest to me that no matter what the Bevyxxa group should have less events (based on phase 2 proven results of anticoagulative effects). It is still encouraging that the medication had no increased bleeding. But these results may explain lack of adoption. It is also likely because many patients will not end up in the hospital for 35-42 days on average, so there may be a lack of overall demand. I am also normally weary of studies reporting composite endpoints, as a single outcome in the overall endpoint can often be a driver of difference, but all of the endpoints in the composite are of similar magnitude and relate to VTE/embolic events. Important, however, and expected, was the study protocol which was double blinded and double dummy indicating that patients receiving Bevyxxa also received "sham-injections." Another issue is drug interactions with other medications that inhibit p-gp (a protein involved in efflux), but we will not dive into the specifics - just that these medications are not rare in hospitals.
I think all things that have been laid out considered, we can see why Bevyxxa has not taken off in practice; even though it may have some real benefits; trial interpretation is muddied, and benefits may not be significant enough to outweigh costs. I believe Portola's model using centers of excellence will serve to benefit Bevyxxa's near future. The addition of a standardized protocol/procedure for appropriateness of use will take away some perceived risk for clinicians while sacrificing present sales ("sacrificing" being a stretch as they were not much to begin with and have still grown since). I definitely see potential for benefit in a medication with a positive side effect profile and possible increased efficacy with ease of administration - as long as there are no drug interactions in the patient's profile.
Next in Portola's pipeline is the Syk/Jak inhibitor Cerdulatinib. Syk is the spleen-tyrosine kinase protein and Jak is the Janus kinase protein. These proteins are often implicated in auto-immune/inflammatory diseases and blood cancers. Many other companies are currently targeting these proteins for other indications. One, as a side note, that I can think of is Pfizer's (PFE) Jak inhibitor in phase 2/3 studies for atopic dermatitis - which seems successful so far.
Portola indicates that the molecule is currently underway enrolling patients in a phase 2a study for relapsed/refractory B-cell malignancies, and is being studied in various T-cell (B/T-cells are types of white blood cells) cancers that have failed/not responded to other therapies. Syk is a protein involved in B-cell and B-cell receptor signaling, while Jak is involved with cytokine signaling causing proliferation and survival of blood and immune cells. We know, specifically, that Syk is involved in activation/differentiation/survival of B cells and the functioning of T cells. In the complete absence of Syk the survival of B cells will be decreased. The hope is that the Cerdulatinib molecule would achieve the same effect in the event that B cells are overpopulating and causing cancer. Jak mutations are linked to many B and T cell leukemias/lymphomas, and inhibition of mutated/overactive protein function aims to reduce survival of malignant cells. Cerdulatinib has been given orphan drug designation (on 9/25/18) for Peripheral T-cell Lymphoma, and PTLA is also pursuing other small molecules for selective inhibition of Syk. But Peripheral T-cell lymphoma is not B cell-based which is why, in my opinion, dual inhibition with Jak/Syk may be working out for PTLA. It seems that Jak and Syk inhibition may have complementary effects on both B and T cell cancers, based on info above. According to the Leukemia and Lymphoma Society, patients with PTCL often have poor treatment outcomes with traditional chemotherapy, which has sparked a lot of research into targeted therapies.
To date, Portola has released some promising info for Cerdulatinib. On 12/3/18, a press release was published:
'The ongoing Phase 2a cerdulatinib data continue to demonstrate meaningful clinical activity, with particularly compelling complete response rates among AITL patients and the potential for durable response among patients with PTCL and CTCL,' said Steven M. Horwitz, M.D., Memorial Sloan Kettering Center, and study investigator. 'Like many others with T-cell malignancies, the patients in this study have failed prior therapies and are in need of additional treatment options. It is encouraging to see such a strong early signal in both disease response and quality-of-life measures, such as itching.'
A very bullish outlook on the medication in early phase 2 studies, and interim results at that. AITL, PTCL, and CTCL are Non-Hodgkin's Lymphoma subtypes that are quite rare can be difficult to treat and are often aggressive, according to the Leukemia and Lymphoma Society. The study is published on clinicaltrials.gov and pubmed. While no specific results have been uploaded yet, the company reported that so far: Patients with AITL had a 50% complete response, 27% of patients with PTCL had a complete response, and in CTCL, only 7% achieved complete response, with 19% achieving partial response, but 48% with a reduction of at least 50% in skin lesions. The drug also, "achieved rapid reduction in severe itching - a common and often serious condition associated with CTCL. The CTCL study continues to enroll patients, and the others continue with follow-up. Important to consider is the absolute numbers of patients in this study are quite low to date, which leaves a lot of room or chance and error. Just as important, however, is the target population which is those who have failed other therapies, or are refractory to treatment. This medication represents a last ditch effort for them.
It is clear that Portola has enough money to get through at least FY2019 and part of FY 2020. Even in the presence of $125 million in debt, with the current revenue growth rate attributable to Andexxa, expanding availability, and comfortability with the product, we should expect the company to reach profitability soon. At the current price in the low $27 range, and market capitalization of $1.8 billion, it seems Portola may be of value to investors. The 52-week high is $45.68 and all-time high occurring in 2017 ($66.04), when there was maximum Andexxa and Bevyxxa hype. I think there is precedent for room for equity growth, even though the past is not a good indicator of future.
Taking the Andexxa revenue growth with the potential that standardized recommendations for use of Bevyxxa could bring to the table, along the promising results of Phase 2 studies for Cerdulatinib, we have a speculative stock that already has revenue and some stability. It is still interesting to note the short float is in the 22% range. So, should none of this happen, and we assume Bevyxxa continues to disappoint in sales with minimal success from Cerdulatinib in future trials, I still think Andexxa is valuable to any pharmaceutical company's portfolio, and Portola could become a candidate for a buy-out.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. 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.
Additional disclosure: I am not a licensed professional in the financial industry, nor do I have a formal education or training in anything but pharmacy. All views are my own and do not represent the views/opinions of any affiliates or employers.