Flexion Therapeutics (NASDAQ:FLXN) presented its Q3 quarterly update on March 12, 2020 (As we tracked on our catalyst calendar on Ampbioresearch.com). Therefore, we took an updated look at the company as we decided whether to change our FLXN positions in our portfolios ahead of the quarterly call. This article is not intended to be a general review of FLXN. For such a review you can visit the company’s web site, including its more recent corporate slide deck. You can also visit the ZILRETTA/knee osteoarthritis (OA) pain page on Ampbioresearch.com for information on ZILRETTA for knee OA, including a list of ZILRETTA publications with links.
In our view, for FLXN at this point, it comes down to how well the company commercializes ZILRETTA in the osteoarthritis knee pain market. ZILRETTA is a long-acting extended release therapy that was first approved by the FDA in October 2017 for the treatment of knee pain associated with osteoarthritis. An sNDA (updated label) was approved for ZILRETTA late last year related to repeat administration. With over 6 million steroid injections annually to treat pain associated with osteoarthritis of the knee, the market opportunity for ZILRETTA is very large. Whether FLXN can tap into enough of this large market to drive strong share price gains remains to be seen. In this article, we attempt to identify the reasons that FLXN ZILRETTA sales may be flattening out and what could drive share increase in the future.
FLXN’s stock price is primarily driven by revenue and clinical performance of ZILRETTA. Therefore, we were excited in mid-2019 about FLXN when sales of ZILRETTA showed good quarterly growth (See graph below). Unfortunately, sales appeared to flatten in the second half of the year. The quarterly revenue growth rates were 60% in Q2, 28% in Q3, and 9% in Q4 by our calculations. Furthermore, FLXN management guided to flat sales in Q1 ’20 vs. Q4 ’19, which should be considered in view of first quarter revenue being flat or down for biopharma sales in many cases, for various reasons including consumers’ reluctance to spend in Q1 when they will pay out of pocket deductibles. FLXN management boasts about its 2019 vs. 2018 revenue growth and points out that hyaluronic acid injection sales are typically down 10% in first quarters annually.
Source:(Amp Bioresearch FLXN ZILRETTA Page)
The question for investors is whether this apparent revenue plateau justifies the current fairly low stock price, or whether the projected 2020 revenue growth and/or the large market potential, provides a FLXN stock price growth opportunity in 2020 and beyond. The company has guided for 2020 ZILRETTA revenue of between $120 and $135 million, which would be between 65% and 85% year over year growth from the $73M of 2019, which was 220% higher than 2018. However, using the Q4 run rate of about $24 million, this would be annual growth of between 25% and 41%, compared to a 36% annual growth rate using 4X the Q3 to Q4 2019 growth. Thus, the company predicts continued growth similar to or slightly less than the Q3/Q4 rate, which arguably has not impressed the market given stock market performance, likely because of ZILRETTA’s current relatively small revenue base for a therapeutic. FLXN’s stock performance in 2020 thus far seems consistent with this viewpoint and the more recent macro downward pressures. In this article, we try to determine whether there is upside in FLXN’s guidance and stock performance for 2020 and beyond.
Let’s start with the opportunity. The market for pain injections for osteoarthritis is large. FLXN cites IQVIA data (see slide below) that indicates that the market is around 5M insured U.S. patients per year. Furthermore, the data FLXN presents indicates that 4.5M of these patients receive steroid injections for a total of around 7 million steroid injections per year. At $300/injection wholesale price (assumed to be about of $600 current cost to end user, see below), that total market opportunity appears to be in the billions for ZILRETTA, and even more if one uses FLXN’s report that a WAC of $570 is justified (See FLXN March 2020 Corp slide deck slide 19).
Source – FLXN March 2020 corporate slide deck
Furthermore, diabetics with OA of the knee is a significant market. A recent CDC publication indicates that 17 percent of Americans between 45 and 64 years of age, and 25 percent of Americans 65 years and older have diabetes. FLXN’s CEO on the recent Raymond James investor call indicated that about 20% of OA knee patients are diabetics. Therefore, if we use FLXN’s numbers of around 7M steroid injections per year, around 1.4 million of those are for diabetics. At $300 wholesale per ZILRETTA injection, that’s a submarket of over $400M, and almost twice that if we use the wholesale price (i.e. WAC) that FLXN indicates is supported.
With this large market opportunity, why has FLXN not been able to capture, and guide to even more market adoption of ZILRETTA in the knee OA pain market? A number of reasons have been discussed on investor calls. One issue for ZILRETTA is competition from generic steroid injections and other choices for injections for knee OA pain. There are a growing number of options for injections to treat joint pain such as hyaluronic acid (HA), platelet-rich plasma (PRP), and placental tissue matrix (PTM) (Cleveland Clinic). We are not aware of a head-to-head trial of ZILRETTA with any of these other injectables. However, a recent review article (Bisicchia and Tudisco, Clin Cases Miner Bone Metab. 2017 May-Aug; 14(2): 182-185) that analyzed prior primary and meta-analysis of a number of hyaluronic acid (HA) trials, concluded as follows: “Conflicting results have been reported in clinical studies and meta-analysis on the efficacy and safety of HA. Guidelines are controversial and “uncertain” recommendations are provided in most of the cases due to inconclusive evidence in literature. However, HA does not seem to have significantly higher side effects when compared to saline or CSs injections, and provides better medium-term control of symptoms in patients with mild to moderate knee osteoarthritis. More studies are needed to better clarify the controversies on this topic, along with a homogeneous methodology in study design, and collection, analysis, and interpretation of data.”
Thus, at least from these authors, HA may only be marginally better than immediate release corticosteroids (CSs). In fact, the 2013 Guidelines of the American Academy of Orthopaedic Surgeons noted that published studies showed a statistically significant but not clinically effective response to HA injections, and do not recommend using HA injections to treat knee pain (Guidelines). What about PRP? According to “Your Guide to Injections for Knee Osteoarthritis” on everydayhealth.com, there is not enough evidence to show a benefit of PRP over HA. However, there are peer-reviewed publications such as Guvendi et al (2018), that conclude that there is a long-term benefit of PRP over CSs, although there are limitations to this study, (e.g. it was not a blinded study). Kavadar et al. published another study that showed long-term benefit of PRP, although the benefit appeared better with multiple injections, but there was no placebo control group in this study (Kavadar et al. 2015).
We do intend to provide a thorough scientific review of data on options for injections to treat knee pain. However, from what we found, CS appears to still be the first choice in OA knee pain injections, but HA and PRP are competitive options. The authors of the HA review cited above call CSs the “gold standard” for injection into a joint to relieve pain because it is well established over decades for having short-term pain benefits. The Cleveland Clinic page on injections for joint pain cited above call CSs “the first line of defense.” And this 2013 article looking at cost/benefit of different knee treatment options, confirmed steroid injections as the standard of care. Furthermore, it appears to be common to give the same patient a CS injection plus other pain injections at later times (See e.g. everydayhealth.com article cited above). Therefore, steroid injection still appears to be a well-accepted initial injection for the treatment of pain associated with OA of the knee. However, there is likely strong competition when it comes to marketing a therapy like ZILRETTA, an extended release CS, for better long-term efficacy compared to immediate release CS. Thus, one factor affecting ZILRETTA’s adoption is likely competing therapies that provide benefits over immediate release steroids. Nevertheless, if FLXN’s sales force can convince orthopedic MDs that ZILRETTA has enough advantages over immediate release CS and the economics of using ZILRETTA make sense for their practice, it still seems possible that a significant portion of this large market could be available with ZILRETTA as a replacement for immediate release CSs, especially if orthopedics felt that other options such as PRP could be used if pain returns after 1 or multiple ZILRETTA injections.
The ZILRETTA efficacy data in the label shows separation from immediate release steroids for a single administration.
Source: ZILRETTA label (package insert)
It is unfortunate that we don’t see more separation in efficacy over time in the ZILRETTA arm of the study in the label to further substantiate its extended release formulation. However, FLXN sales reps can provide orthopedics with several ZILRETTA publications that provide more convincing data of the benefits of ZILRETTA over immediate release CSs. For example, the Conaghan et al. JB&JS 2018 publication of the phase 3 results, which shows statistically significant improvements across pain, stiffness, physical function, and quality of life for 4, 8 and 12 week time points (See Table IV from that publication below).
Source: Conaghan et al. JB&JS 2018 publication of the phase 3 results
Furthermore, data on these endpoints is even more convincing on a post-hoc analysis done on the subset of patients in the phase 3 study participants with unilateral knee OA (Langworthy et al.). Thus, it appears that there is substantial data that the FLXN sales team can use to show that ZILRETTA provides superior efficacy to immediate release CSs to try to convince orthopedics that ZILRETTA should be used instead of immediate release CSs as a first injection therapy.
Another issue for FLXN with respect to its inability to grab more market share is the fact that ZILRETTA’s first approved indication was only for a single administration. Steroid injections are often re-administered to a patient, and FLXN felt their sales force was at a disadvantage with the initial label that was limited to single administration. Fortunately, in Q4 2019 the FDA approved FLXN’s sNDA for an updated label for repeated use of ZILRETTA. However, the approved label language doesn’t seem that convincing for repeat use. Here is FLXN’s press release, which notes the change in intended use language as follows:
“Removal of language which stated that ZILRETTA was “not intended for repeat administration.” The updated label states that the “efficacy and safety of repeat administration of ZILRETTA have not been demonstrated.” “
Here’s the actual label, intended use section:
Source: ZILRETTA label (package insert)
Furthermore, the safety language on the label for repeat administration doesn’t look that helpful either
Source: ZILRETTA label (package insert)
The CEO of FLXN clarified at the recent Raymond James Investors Conference that he feels it is now a much better situation for his sales team post sNDA approval, since they can present to, and specifically discuss the publication of the results from their repeat administration publication with potential prescribers. Prior to the expanded label approval, they could not discuss the repeat administration results with prescribers, but only refer them to the FLXN scientific affairs team if they had questions. Therefore, hopefully the FLXN sales force can use the repeat administration publication to drive more sales. With respect to repeat administration, ZILRETTA appears to be effective on a second administration 12 to 24 weeks after an initial administration based on FLXN’s repeat administration trial results. Although efficacy appears to decrease slightly upon a second administration of ZILRETTA, the data still looks very convincing that ZILRETTA remains very active on a second administration (see graph below from Spitzer et al. Rheumatology and Therapy 2019). Thus, FLXN may be able to convince orthopedics to use ZILRETTA initially instead of immediate release CS, and repeat ZILRETTA administration at 3-5 months to get more relief, and then if the pain returns, potentially switch to another type of injection such as PRP or opt for knee replacement surgery.
Source: Spitzer et al. Rheumatology and Therapy 2019
It appears that repeat administration efficacy data is strong enough to confirm that ZILRETTA is effective on a second administration. The questions are whether the data is strong enough to change prescribing habits and compete with alternative long-term pain relief options, and overcome possible practice economics issues (see below) to drive substantially more sales of ZILRETTA.
Another advantage of ZILRETTA is that it is particularly well differentiated over immediate release CSs for diabetics who receive a CS injection in a painful OA knee to reduce pain. An immediate release CS injection in the knee brings a spike in glucose. This spike is significantly reduced when ZILRETTA is used instead of immediate release CS (See slide below).
Source – FLXN March 2020 corporate slide deck and FIG. 2 of Russell et al. 2018.
It is noteworthy that despite the title of that slide from FLXN, that graph is not in the label. We believe FLXN is referring to the following statement in the ZILRETTA label:
Despite the minimal coverage for diabetics in the ZILRETTA label from our viewpoint, FLXN’s commercial team should be able to leverage the published data for diabetic patients to drive more sales in this subsegment of the market. For example, ZILRETTA showed an increase of glucose of 155.24 before administration to 163.41 post-administration, which was significantly less than the increase seen with immediate release CS of 161.71 to 198.78 (See FIG. 3 of Russell et al. 2018). With respect to the diabetic market however, in the recent Raymond Jones call, the FLXN CEO was quick to point out that the company did not want to pigeon hole the market for ZILRETTA into diabetic patients.
Another reason that has been brought up on investor calls that potentially is holding back more rapid and extensive market adoption is that the cost and reimbursement situation for ZILRETTA is not attractive for providers such as orthopedic practices and/or hospitals. This is not an area where we have deep understanding and in general actual net revenue per injection information is not publicly available, but as far as we can tell this might be a key issue holding back ZILRETTA adoption. The retail price of ZILRETTA as best we can tell appears to be around $600 (Drugs.com, Oklahoma Health Care Authority 2018 (Packet Contents for DUR Board Packet – January 9, 2019) annual review of ZILRETTA (pg. 93) (download available at www.okhca.org), and webmd.com), which appears consistent with the $604 average selling price we find in CMS’s reimbursement table effective January 1, 2020 based on ZILRETTA’s Jcode (J3304) (See FLXN corp deck slide 20) ($18.88 per mg and an administration of ZILRETTA is 32 mg), which FLXN indicates is about 104% of the ASP (FLXN March 2020 corporate slide deck). This price seems similar to the price of HA (Brett 2016 Editorial) or PRP injections (Everyday Health). The average retail price we find for Kenalog-40 an immediate release triamcinolone, the CS in ZILRETTA, is about $12.00 (See e.g. Oklahoma Health link above and drugs.com). For a person paying out of pocket, this almost $600 difference may not be worth the increased efficacy and even safety if they are a diabetic. Even for a payor, this difference might require a special situation, such as diabetes (See Oklahoma Health link page 93, “A patient-specific, clinically significant reason why the member cannot use Kenalog-40… must be provided”).
Since these knee pain injections are administered by providers at their outpatient facilities and many of the patients receiving injections are on private insurance or Medicaid, the situation is more complex. FLXN has reported good commercial insurance coverage and that Medicaid will reimburse providers, since there is a J code and ZILRETTA is delivered on label and for an indication that is medically necessary and reasonable (See FLXN March 2020 corporate slides 1-21). The more specific issue might be in the amount that providers such as orthopedic practices and hospitals are reimbursed, especially for Medicaid patients, compared to what they are paying for ZILRETTA, and how that compares to their reimbursement and cost for immediate release CSs. Plus, payors, such as the Oklahoma Health Care Authority, may be reluctant to pay such a high price premium without special circumstances, such as diabetes as noted above.
FLXN appears to acknowledge that there is an issue(s) that they need to overcome to reach the full market potential for ZILRETTA. The CEO indicated on that same Raymond James call that they still want to hire a Chief Commercial Officer (CCO) who can help FLXN figure out how to significantly increase its market share. In fact, FLXN announced the hiring of a new CCO this week. Thus, either FLXN doesn’t understand the reason that there is not more market adoption, or understands the reason(s) but doesn’t know how to fix it. Until we see some quarterly revenue that shows a more significant increase than current guidance, we do not predict that FLXN will exceed its guidance for 2020 and find it difficult to predict revenue in future years without seeing 2020’s actual revenue numbers.
The vast majority of the value of FLXN currently, resides in ZILRETTA. However, FLXN has some other assets as well (See our ampbioresearch.com Corporate profile database slide below), but they are much earlier phase assets. Furthermore, FLXN has some mid-late stage ZILRETTA trials, for example to expand the label into shoulder OA pain (see below).
Source: (Ampbioresearch.com FLXN corporate database page)
If FLXN’s commercial team can have success tapping into the OA pain market for ZILRETTA, it does not take a detailed valuation to see that FLXN’s valuation and stock price can increase substantially over its current ~$250M market cap and enterprise value of roughly $300M (about $190M debt and about $135M cash). If one uses even a 5X revenue multiplier, a fairly conservative multiplier for a biopharmaceutical company, and FLXN is able to reach its $120M low end guidance this year, there appears to be upside in the stock price given the recent market correction. In the coming few years if FLXN can achieve $200M in sales, that would warrant a $1B market cap. Of course, there are a lot of factors that go into the revenue multiple in a stock price and increased revenue growth and apparent maximum revenue compared to cost of goods, R&D expenses supporting revenue growth, and especially the selling portion of SG&A expense would dictate FLXN valuation on a long-term basis. If FLXN could find a way to tap into much more of this large knee OA market, based on increased revenue and an increased multiplier because of an increased growth rate, it is not hard to get to double its current valuation and stock price. By our calculations using a rough free cash flow model (data not shown), it also appears that FLXN is undervalued currently, given its recent significant stock price drop, if it can hit its 2020 projections despite the pandemic.
Using Q4 2019 numbers, FLXN had about $135M cash/equivalents/securities and a burn rate of about $35M/quarter. Thus, FLXN ended Q4, 2019 with about 1 year of cash and a growing revenue stream that could decrease the burn rate depending on 2020 expense management. Although on their most recent call, FLXN management again noted that they projected that they would not need to raise more cash before they became profitable, this seems unlikely under the current 2020 revenue guidance and quarterly spend especially as they indicated an increasing R&D spend this year. However, it does not appear that a dilution event is imminent at this time, but probably will occur later in 2020.
In sum, we currently hold a low percent position in FLXN in a number of our funds, and will continue to monitor FLXN closely. The macro pandemic news has really hit FLXN hard the past week, which has quickly created a valuation opportunity when the markets return to normal. We are interested in the readout in their ongoing OA shoulder pain phase 2 trial of ZILRETTA, but that is not scheduled to read out until the first half of 2021. However, we are most interested to see if FLXN, especially with a new Chief Commercial Officer, can find a way to more significantly and rapidly increase its market share in this large market opportunity, or execute on the 2020 guidance and convince the market that peak sales of ~$400 million within 10 years is achievable despite the challenges of a highly competitive marketplace. Given the recent correction based on the macro downturn from the pandemic, at its current market cap, FLXN appears undervalued. Furthermore, although we do not invest hoping for an acquisition, it would be a nice surprise for shareholders if an entity with an effective commercial team in the pain/orthopedic space is willing to pay a premium on the current depressed FLXN stock price based on the commercial potential of ZILRETTA. Thus, we will continue to closely monitor FLXN’s commercial and clinical progress in 2020.
Quick Note: Thanks for taking the time to read our article – hope it was helpful and useful. If it was beneficial to you in any way please consider hitting that ‘follow’ button to stay updated of our future analysis. Also wanted to note that these articles, including a version that we published before the FLXN quarterly report-out, are available in advance to our free community over at Ampbioresearch.com, including educational videos and more. If you have any questions or comments we would love to hear them!
Disclosure: I am/we are long FLXN. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.