Collegium Pharmaceutical (COLL): 100% TSR Pharma with a call option worth 80-100% TSR more
Current Price: ~$29-30 | 36-month Target Price: ~$48, TSR: 100% | COLL
Join the mailing list for future blogs: Click here
Model: Available to download here
Overlooked pharma company which can generate enough cash to purchase all
shares within the next 6-7 years at current price levels
Company & Industry Overview
Collegium pharmaceuticals was founded
by Michael Heffernan who stepped down as CEO mid 2018 when Joe Ciaffoni took
over. He has since stepped down and Michael Heffernan who stayed on Chairman
has become interim CEO. Michael will now be succeeded by Vikram Karnani, who
grew Horizon sales from $300m to $4b in 9 years before acquisition by Amgen.
Collegium markets extensive release opioid
pain medications. Important
note: Collegium was named, but subsequently dismissed with prejudice of any
opioid related litigation against Purdue Pharma, distributors and other generic
pharmaceutical companies. Its two main pain growth medications include
(1&2): -
1. Xtampza ER (an abuse deterrent patented oxycodone-based opioid, very similar to Purdue Pharma’s OxyContin with much more abuse deterrent technologies).
2. Through acquisitions, it owns belbuca (the only buprenorphine reference listed drug/ branded drug that treats chronic pain). Belbuca isn’t a capsule or tablet, rather a strip to be attached in the mouth which dissolves. Other Buprenorphine applications including drugs like Suboxone, Sublocade, Brixadi, Zubsolv treat Opioid misuse disorder by including naloxone or other related APIs.
3. Other drugs include Nucynta IR, Nucynta ER and Symproic. Former 2 will have generic entrants by 2026 and latter is a drug substance and note product. Negative growth expected from Nucyna assets, and no growth from Symproic
Since 2015, opioid prescriptions have
fallen. In 2018, 189 million opioid prescriptions were written, this was 140
million in 2023, the same level to 2020. However, these are overwritten by
massive controversies such as Oxycontin, which Xtampza ER was made to override.
More so, 13.6 million generic and branded prescriptions were written in 2023,
and 2.6 million branded ER (Extended release) prescriptions in 2022.
What is ER and IR? ER is extended
release, which means that chemical releases slowly, and IR is immediate like a
paracetamol, instant. Victims of chronic pain would usually take 3-4 IRs vs 1-2
ERs and have less chemical volatility. 85% of ER users have switched from IR
but not vice-verca.
Why are we buying a pharma company
with no R&D?
Overridden from predeceasing
controversies of other industry players (not collegium), makes a company such
as collegium massively overlooked. Put simply, Collegium has superior economics than most
pharma players that generates superior cash flows in proportion to revenue &
market capitalization + distributes that cash flow in share buybacks.
Thesis 1: Market participants are
ignoring the runway and obvious differentiable molecular structures and
embedded technologies for three key growth assets (Xtampza ER, Belbuca, Jornay
PM)
1. Xtampza ER
In 2017, Oxycontin OP (Oxycontin
version with some abuse deterrent technologies developed after Xtampza ER) did
$1.7b in net sales, the year before it did 2.1, Xtampza is ~180m in 2023 end.
Post Purdue Pharma LP’s bankruptcy
and the Sackler family paying a $4b+ fine, and a famous Netflix series, Purdue
Pharma LP actively does not actively promote Oxycontin. There are only two
branded oxycodone extensive release opioids, Xtampza ER + Oxycontin. Generic
versions of an oxycodone ER are only authorised generics of Oxycontin by Purdue
Pharma. By gross sales
(before rebates and discounts), Oxycontin branded is still 60%+ market share vs
sub 40% for Xtampza as of Q2 2024. They have increased market share
above 40% in Q3.
|
Fig 1: The problem with
OxyContin OP (An abuse “deterrent” drug) |
|
|
Except Xtampza ER, all labelling has
“crushing, dissolving or chewing can cause rapid release and absorption of a
potentially fatal dose of the active drug”. This is different to API drugs
such as: -
|
Drug |
Company |
Problem |
|
Hysingla |
Purdue Pharma |
Hydrocodone, once a day (not sufficient) |
|
Embeda |
Pfizer |
Morphine and naltrexone, naltrexone is an opioid antagonist that
can be sprinkled on food |
|
MorphaBond ER |
Inspirion Delivery Technology |
Morphine |
|
Arymo ER |
Egalet |
Morphine |
|
Fig 2: DETERx solves this
problem |
 |
|
|
Patent No. |
Expiry |
|
1 |
8557291 |
21/03/2025 |
|
2 |
7399488 |
24/03/2025 |
|
3 |
7771707 |
24/03/2025 |
|
4 |
8449909 |
24/03/2025 |
|
5 |
8758813 |
06/10/2025 |
|
6 |
9682075 |
12/10/2030 |
|
7 |
10004729 |
12/10/2030 |
|
8 |
10668060 |
12/10/2030 |
|
9 |
9737530 |
02/09/2036 |
|
10 |
9968598 |
02/09/2036 |
|
11 |
10188644 |
02/09/2036 |
|
12 |
10646485 |
02/09/2036 |
Each patent has an element of its
deterrent technology. Notorious generic (+ innovator) Teva pharmaceutical tried
infringing upon patents by filing a ANDA to Xtampza ER but failed and the
settlement permitted ANDA exclusivity to only Teva post 2032. As per external
studies, there is a 700m peak potential by a study in 2013 study pre-launch (my
model takes 320 by 2032, less than 50% of the survey that based their internal
efforts pre-launch and post-launch). I believe I am well under a margin of
safety for my projections given the following reasons: -
1.
Certain market participants may be worried about falling
prescriber data for Xtampza : -
|
Period |
Unique
Prescribers (Source Quarterly presentations) |
|
Q2, 2024 |
16,600 |
|
Q1, 2024 |
16,300 |
|
Q4, 2023 |
17,000 |
|
Q3, 2023 |
17,700 |
|
Q2, 2023 |
18,100 |
|
Q1, 2023 |
19,000 |
|
Q4, 2022 |
19,000 |
|
Q3, 2022 |
19,300 |
|
Q2, 2022 |
19,200 |
|
Q1, 2022 |
18,700 |
Some definition time (GtN, if you know this,
ignore it): Under
the Medicaid act, government rebates are applicable to most drugs based on
coverage. The difference between WAC (Wholesaler average cost) and price it
gives to retailer, if at a discount is recoverable (industry norm). Amongst
many other things, these are called rebates, returns and discounts (also
includes co-pay as drug sales are heavily covered by 3rd party
coverages). Hence if the gross price of a drug being $100, net sales are
usually $40 - $60, 60% GtN Lleading to $40 and 40% GtN leading to $60. The deduction from Gross to net
is called GtN.
Xtampza’s GtN in 2022 was 69.3%, it
successfully fell to 59.6% in in 2023. It plans to take this to 55-57% this
year. Gross pricing was only mid-single digit that year as per mgmt. due to
inflation related provisions within Medicare (historically Jan 1 is 9.9% for
Xtampza) plus protecting against future rebates from IRA. Even if the average
prescriber base went down 6%, its not an indicator of falling reliability of
the drug. The pricing and incentives massively changed as collegium focused on
less capital intensive and more margin accretive growth. Also falling
prescribers doesn’t necessarily mean falling prescriptions, nor does it mean
losing market share, as Xtampza has gained market share. In fact this is mostly
from prescriber data that doesn’t include commercial coverage.
Lastly, Collegium guides 66% opioid
ER pain specialists reach in 2024 using their 110 numbered sales team (for the
whole company). Conversely, Purdue pharma at its peak used 351 dedicated only
to oxycontin. Lastly, Xtampza had 166,400 prescriptions in Q1 2021 (company not
reported since) of 2.9 million total branded ER prescriptions, it has grown
most quarters while total prescriptions have fallen due to many opioid
companies taking reputation hits from lawsuit settlements. There is a sufficient runway
with no red flags as such till 2032.
|
Fig 3: Xtampza ER revenue |
 |
2. Belbuca
Summary: Belbuca has a settlement
with Teva for a generic in 2027, but that’s not an end of world scenario
Neither is the market nor am I in
doubt of growth for Belbuca. This is a drug that has grown prescription count every
quarter on quarter ignoring the
donut whole effect of Medicare, where Q1s are usually the worst
quarters. In addition, it has won a 2 million Medicare Lives contract with 8
million commercial lives with Xtampza and exiting a 8 million high GtN
coverage.
It’s an extremely unique drug, which
uses the least amount of buprenorphine amongst the patented RLDs (Reference
listed drug title given to innovators) and is the only chronic pain medication
using buprenorphine (Class 3 narcotic vs Class 2). Other RLDs use this API to
treat opioid addiction. Double
digit growth is highly probable with stable industry average GtNs. Till when is
the question?
Teva had a settlement with BDSI (a
once former company acquired by Collegium) for launching a generic 2027 Q1. Belbuca
has three patents: -
|
|
Patent No. |
Expiry |
|
1 |
8147866 |
23/07/2027 |
|
2 |
9655843 |
23/07/2027 |
|
3 |
9901539 |
21/12/2032 |
Producing dosages of belbuca without
infringing the third patent is difficult which makes economic considerations on
investment sizes, as generic production without scale isn’t highly profitable.
“Everyone underestimates the power of
incentives” Charlie Munger
“Teva will stop producing some older
generic drugs and reduce the number of new generics it develops” Teva CEO,
Richard Francis 18 May 2023
Teva is going through a major
deleveraging cycle post $20b toxic debt worrying shareholders. Furthermore, for
all states, Teva pharmaceuticals also must pay $4.2b nationwide over 18 years for its role in the opioid
epidemic. Teva also dropped its first filers exclusivity (FDA gives
180 day generic exclusivity), which allows Belbuca to launch its own generic as
a response immediately as well. Alvogen lost its case and has settled on
launching post 2032. The chemo group is still on the FDA stage of approval
after receiving its fourth CRL (notice issued by FDA of non-approval in current
form). This is a testament that developing all doses of Belbuca is difficult
without violating patent 9901539.
As per my base case, I have assumed a
13% compounded price decline + 60% volume for the branded version from 2027,
then 25% every year to presume 90% volume decline over 5 years. The remaining demand
is filled 50-50 between Teva and Collegium for generics. I have taken the
appropriate GM% assumptions for it as well. However, valuation will cover a scenario that Teva
doesn’t pursue this generic that gives more meaningful growth to Belbuca, and a
lucrative call option.
|
Fig 4: Belbuca Revenue
(Pre-22 revenue from BDSI 10k, 22 revenue estimation) |
 |
3. Jornay PM
Summary: Clear patent protection till
Q1, 2032 + clear differentiation of product + long runway of growth for Jornay
PM (drug for ADHD; just finished acquisition)
Jornay PM has 16 patents all expiring
on 23rd March 2032. Jornay PM’s growth runway currently has been in
excess of 60% a year reaching $100m, and I expect it to be 25% till $125, and
15% thereafter. The reason of my confidence is for 2 things: -
1.
Mgmt has guided it being the highest top line growth asset
2.
Other namely stimulants include adhansia XR, aptensio XR,
concerta, cotempla XR-ODT, metadate CD, Quillichew ER (same API with patents),
Quillivant XR, Relexxii, Ritalin LA are taken in the morning. Jornay PM uses
delexis, a patented platform that delays initial release of the drug for up to
10 hours. This reduces the need for a child to remember to take its medicine
(Reduced cognitive load) + wakes up fresh + plus required only 1 dosing
Jornay PM is 80% paediatric, with
collegium exploring adult synergies. 6.8 million Children (Source: CDC),
above age 6, suffer from ADHD. 52 weeks a year or 260 working days, i.e. 260
pills per child. To reach 360 million USD, or 750 million gross revenue, jornay
pm at a mid single digit price increase would have to sell at $24 vs $16 today,
and sell 31.25 million pills, or an average to 120,000 children each year at
100% paediatric coverage. That is a 1-2% penetration rate to achieve by 2031.
While mgmt. doesn’t give peak revenue
guidance they have valued the acquisition as a $635m intangible asset
acquisition. In comparison, I am at $500. This would imply a compounded average top line growth of
15%+ vs mine at 14%+ with peak revenue at $340m vs $315 for mine, a FCF to
sales margin of 55% vs mine at 48%, which also means an EBITDA margin of 60% vs
mine at 55%.
Thesis 2: Above top line growth, this
is a well-oiled machine that the market is ignoring as it isn’t an indefinite
asset with R&D
|
Fig 5: BDSI Non R&D
Operating exp % sales |
Fig 6: COLL Non R&D
Operating exp % sales |
 |
 |
Collegium invested through and
successfully rolled over all operational requirements of BDSI (Owner of
Belbuca) within its existing infrastructure. Fig 5 vs Fig 6 will also show BDSI, a company with the same
corporate philosophy, was consistently worse off than collegium in efficiency
of resource allocation.
|
Fig 7: Index 1 vs COLL non-R&D op exp %
sales |
Fig 8: Index 2 vs COLL
non-R&D op exp % sales |
 |
 |
Fig 7 compares an index comprised of US
branded pharma companies that are EBITDA positive and revenue between $200m and
$350m vs collegium at similar revenue scale from 2018 to 2021 to depict collegium has been efficient at
this scale with allocation of resources.
Fig 8 compares an index comprised of
US branded pharma companies that are EBITDA positive and revenue above $350m vs
collegium at similar revenue scale from 2022 onwards to depict collegium has been efficient at
this scale with allocation of resources as well.
How and why?
After reviewing employee reviews on portals,
it seems that the culture is competitive. Paraphrased from an employee’s words,
“targets are unachievable that are set due to performance of few top
employees”. Whether such a competitive culture is the right one is
difficult to answer, but has been sustained over an extended period of time.
Nonetheless, it is a characteristic
of the firm that makes it attractive as margins are efficiently higher, and in my opinion
sustainable and not easily imitable overnight. It also
helps that their assets are unique and in a niche dominated by them (Branded
extensive release).
Thesis 3: Obvious but ignorable
catalysts in form of distributable cash flow
I ran a screener for 170+ US Small
cap pharma names that aren’t 0 revenue. Most hold heavy inventories due to API
shortages generally (110+ days average); Collegium is higher at 150 albeit with
more prescription growth vs average developed assets. Receivable days average
at 90 due to concentration of revenue as 3 major distributors control US drug
distribution. Collegium sits at 115. Collegium has a 240-day working cycle and
is a cash intensive business that wouldn’t have a normalised ROE of 40%+ without
substantial but under 2x net debt ebitda leverage. However with FCF frequently
> 150% of net income and the business not requiring to hold more than 70% of
non-tax cash expenses (240/365 days is 65%, 70% gives good margin of safety), it
gives sufficient room for FCF to be opportunistically returned to shareholders
as depicted in Fig 9 and Fig 10.
|
Fig 9: Within conservative
projections cash balance > cash expense budget |
Fig 10: Free cash flow
(Includes M&A) > Share buyback projections |
 |
 |
The stock has lost some support from lack of buy backs as Collegium has just finished a major acquisition. As Jornay PM begins generating free cash flow alongside Xtampza ER and Belbuca, they have the capabilities to redistribute $850m of buyback on a $975m market cap company over the next 4 years. This alongside repayment/ settlement of convertible notes will close technical shorts (>18% of shares outstanding) that should provide meaningful price action. Most shorts as per my analysis are a results of convertible arbitrage.
Valuation
I haven’t done a comprehensible comparable
analysis for collegium as it will uncharacteristically highlight the stock as relatively
cheaper. Comparison isn’t apples to apples here as a double-digit forward
multiple is attributable to non-terminal assets, which isn’t the case for
collegium yet. I have done a DCF till 2032, last of its patent expiries, post which I have taken a -100%
terminal growth rate. Ordinarily, majority of a DCF value is the
present value of the terminal value. In this case, it’s 0.
|
Fig 11: Base Case DCF
Evolution |
 |
I have done a YoY DCF analysis
shifting present date and discount factor as the years progress to depict an
approximate evolution of share price.
Under a scenario Teva doesn’t produce
a Belbuca generative, I reach the following scenario (without scaling buy
backs) :-
|
Fig 12: No Belbuca generic
case evolution |
 |
Under a scenario where I match
mgmt.’s Jornay PM valuation with projections, the DCF evolution is as below
(without scaling buybacks) :-
|
Fig 13: Mgmt Case for
Jornay PM |
 |
Risks
· Allocation of excess cash flow: Business development, i.e. M&A is a clear prerogative of mgmt. My projections and expectations don’t price this in as that is pure speculation with what, how and when they will acquire.
−
History: Nucynta’s acquisition was a decent acquisition with
high single digit ROICs, Belbuca was a better transaction with double digit
ROICs, and Jornay PM is expected to be an even more accretive transaction to
Belbuca.
−
Mitigants :-
§ Nucynta’s acquisition was
expected as they had a commercial agreement for 3 years before. Hence tried and
tested + expected.
§ BDSI (Belbuca) acquisition
was expected as they had a multi-year collaboration prior as BDSI heavily
relied on collegium drug channels for sales. Tried & Test + expected
§ Jornay PM like acquisition
was guided for 2 years as a non-pain acquisition. Expected
§ I don’t think we will have
a scenario, where there won’t be ample time to react. Rather my valuation exercise depicts a value of
what if rather than what exactly
· New drugs: Vertex’s new drug Suzetrigine is a non-opioid pain medication for peripheral neuropathic pain and moderate to severe pain with a PDUFA for January 2025. While “hyped”, this is more overlapping Nucynta drugs (already facing revenue headwinds and priced in my projections) and maybe Belbuca as a class 3 narcotic (cliff drop post-acquisition already priced in).
−
My thoughts: We saw a drug similar to this in Gabpentin, which
while expected just didn’t work close to opioids. There’s a reason opioid and
pain hasn’t seen “news headline” innovation in the last 20 years. Within our
projections, the only risk it has is to Xtampza ER, which in my opinion is
unsubstantiated today. Xtampza ER is for long-term chronic pain, it is a
completely different TAM.
−
Mitigant: There is risk to Belbuca with this IF it proves showcases
the efficacy it has displayed in trials. However, we have already priced in
Belbuca revenue falling from $280m in its 2026 peak to ≈ $80m due to patent
expiry. If the threat is serious to Belbuca, it is a tailwind for collegium as
this further disincentivises Teva to launch its generic.
· I stress tested IRA (inflation reduction act) increased rebated provision. There isn’t clear evidence collegium drugs have outpaced inflation at net price levels.
Overall, this is a purchase only in
the perspective of a cigar butt on the street with some more puffs that is
cheap. Nothing more nothing less, with a call option from Teva’s decision.
Comments
Post a Comment