Dendreon's Biggest Provenge Problem? Managing Demand
The worry is no longer whether the FDA will approve the drug, but how the company will keep up production.
No later than May 1, 2010 the FDA will make a decision whether to approve Dendreon’s (DNDN) prostate cancer drug Provenge. The FDA, which should have approved the drug in 2007, is very likely to approve Provenge this time around. The FDA has few excuses this time. Provenge has two clearly positive Phase III trials showing a survival advantage, a third Phase III positive once imbalances in favor of the control arm are corrected, and a FDA advisory panel that stated the drug was safe, demonstrated substantial evidence of efficacy, and (by my unofficial count) thought the drug should be approved.
My sense of the investment community -- at least those present at the recent JPMorgan (JPM) Healthcare Conference in January -- is that the “Will they approve it?” question is no longer pertinent for those interested in Dendreon. What’s up next is whether the company can sell it and whether they can manufacture it.
The answer to both is a clear "yes," in my view. Before I explain why, some background is in order:
I first met with company management in 2000. My firm initiated coverage in 2001, and there's only one firm who has longer continuous coverage of the company than ours. Our firm has been positive on Provenge since the first release of data from the 9901 trial way back early last decade. And while predicting Dendreon’s stock price has been the most difficult aspect of this story, anyone following along with the Dendreon stock and option plays in our Biotech Stock Research Model Portfolio have banked significant returns despite all the ups and downs over the years.
The other thing worth knowing is my firm has never issued positive coverage on another active immunotherapy company. This has allowed our clients to miss huge blowups in every other name in this market segment. There are lots of science-geek reasons why. But the biggest reason why is that we never believed other companies could manufacture their drugs at the level the FDA requires of an approved drug (which is a higher standard than required to manufacture for clinical trials).
Manufacturing was one of our biggest concerns in our initial meetings with the company. I was relieved to learn early on it was Dendreon’s biggest concern as well. In fact, Provenge filtered to the top out of a heap of many alternate approaches precisely because Dendreon’s management knew it could be manufactured to FDA specifications. It’s also worth knowing key Dendreon people had gone through Immunex’s Enbrel manufacturing problems -- the problems that lead to Immunex’s acquisition by Amgen (AMGN) for a song.
With that background out of the way, let’s dig into the process.
Provenge is unlike any current cancer drug because it needs a component of the patient prior to manufacturing. Provenge doesn't need a piece of the tumor, which is where most other personalized active immunotherapies fail. Provenge requires a patient’s immune cells -- largely untrained/immature dendritic cells -- to start the manufacturing process. Collecting these cells is simple, requiring the patient visit a blood bank and sit in a chair for one to two hours while they're filtered out of his bloodstream. This isn't special science. In fact, the process utilizes industry-standard equipment already in place at blood banks run by folks like the American Red Cross, New York Blood Centers, and Puget Sound Blood Centers. Dendreon has agreements with all three organizations, by the way, for this step of the process.
Dendritic cells tell the human T-cell immune system what to attack. Prostate cancer cells do a good job of fooling dendritic cells into ignoring the growing tumors. Dendreon solves this problem by removing dendritic cells from this environment, training them outside the patient’s body to recognize cancer cells, and giving these infused cells back to the patient.
Dendreon trains dendritic cells to recognize an antigen called PA2024. This is a recombinant antigen manufactured just like any other recombinant protein. It is a construct of PAP, a prostate cell marker, and GMCSF. The added GMCSF helps the dendritic cells learn to recognize PA2024 more quickly. This is the “secret sauce” of Provenge, is not personalized to the patient, and provides absolutely no challenge in terms of manufacturing.
The Basic Steps
Patients prescribed Provenge will go through the following steps three times in four to six weeks.
1. Visit a blood center near their home to undergo a leukapheresis, the technical term for obtaining their immature dendritic cells.
2. Those cells will be shipped from the blood center to Dendreon’s manufacturing facility.
3. Dendreon takes the next 36-48 hours to train the patient’s immune system cells to recognize the PA2024 antigen.
4. After key potency tests are completed to make sure the training worked, Dendreon ships Provenge back to the prescribing doctor.
5. The patient comes into the doctor’s office and receives Provenge on an outpatient basis via a 60-minute IV infusion.
Two weeks later, the process repeats -- and again two weeks after that. After three treatments, the patient is done and can go on with his life. Clinical trials prove the patient is two to three times more likely to be alive at three years because he received Provenge. Importantly, this significant survival benefit comes with no durable side effects and no serious short-term side effects.
Anyone who has had operational responsibilities for their employer instantly recognizes a ton can go wrong within those simple steps. Dendreon knows this and has been planning on how to handle Provenge manufacturing in the real world for approaching 15 years.
The core of Dendreon’s approach is an enterprise-level software and logistics system called Intellivenge. This software handles everything related to Provenge. Working off of the steps listed above, here is a more detailed analysis of the Provenge manufacturing system.
Over 95% of prostate cancer doctors are aware of Provenge, so Dendreon’s initial problem will be managing demand. The most likely scenario is the company will have too little manufacturing capacity to meet initial demand. More manufacturing will be online next year, capable of handling approaching $2B in annual sales (depending on final pricing). In the meantime, the company will start with basic manufacturing capability at a Morris Plains, New Jersey manufacturing facility.
Demand will be managed at the doctor’s office. Dendreon will designate certain urology clinics and academic centers as what I call “Provenge Authorized Dealers” (PADs). Only a limited number of PADs will be designated to start, which is how Dendreon will limit initial demand. Clinics and hospitals that participated in the Provenge clinical trials will make up the majority of initial PADs. This is part reward for believing in the product when few people did, and part smart resource use as these facilities need less technical support since they are already familiar with the product and manufacturing process. The doctors at each PAD will make a medical decision which patients get Provenge.
A PAD will notify Intellivenge of a patient prescription for Provenge (we’ll call this patient “Eduardo” after a genuinely wonderful early Provenge patient). Intellivenge will help examine hood space (manufacturing capacity) at Dendreon’s manufacturing facilities for an opening. When an opening is found, Intellivenge schedules the manufacturing of Eduardo’s dose. Once this step is complete, Intellivenge works backwards and forwards to schedule all the other steps in the process.
(In fact, a human operator of Intellivenge handles these steps but I found it easier to create this explanation as if Intellivenge was running the show. Since the process would be an operational and logistics nightmare without Intellivenge, this isn’t too far from the truth.)
Intellivenge then schedules an appointment at a blood center close to Eduardo for the leukapheresis. The prescribing doctor, the blood center, and Eduardo are notified by Intellivenge (electronically and via Dendreon’s own call center) about the appointment. The notification includes the initial contact and appointment reminders.
Intellivenge also makes preliminary shipping arrangements with preferred air/ground shipping companies (think FedEx).
The day of the leukapheresis, the blood center confirms to Intellivenge (electronically or phone call to Dendreon’s call center) the procedure was successful. A unique bar code is affixed to Eduardo’s cells, which becomes an important part of a control system insuring Eduardo’s cells get back to him and him only. Intellivenge confirms the previous shipping reservation so Eduardo’s cells are retrieved from the blood center and flown to Jersey for manufacturing.
Barcodes are scanned throughout the process so Intellivenge knows precisely where the bag of Eduardo’s cells is in the process.
Once the cells arrive in Jersey, they are scanned in and the manufacturing process starts. I’ve toured this facility and it is quite amazing in thoughtful design -- if a little underwhelming in gee‑whiz technology. After all, Dendreon’s manufacturing process is “nothing more” than an elegant application of off-the-shelf lab equipment. What’s unique is the coordination of existing logistical and community blood center resources.
First, Eduardo’s cells are washed and tested to make sure they are still alive after shipping. The cells are then separated to get the key immune system cells for training. After testing to make sure they are still viable, they're exposed to the PA2024 antigen for one to two days to be trained. They're washed and tested again after training. The final testing step looks for an immune activation marker called CD54. Upregulation of CD54 within specific boundaries is proven to correlate with survival in Provenge patients and is the test for manufacturing success. The manufacturing success rate in the clinical trials (from leukapheresis to dosing) was 98% -- something no other active immunotherapy company has been able to match and a key component of Provenge’s success.
Intellivenge monitors all these test results and the process of Eduardo’s cells through the Jersey facility. During the process, Intellivenge (electronically and with Dendreon’s call center) confirms an appointment with the prescribing doctor and Eduardo for the Provenge dose infusion. Return shipping to the doctor’s office is also scheduled. Reminders on all of this are sent via Intellivenge.
The shipper picks up the finished product from the Jersey facility and flies it to the doctor. At the point of receipt by the doctor, the doctor is charged for the product. Eduardo comes in, gets a couple of Tylenol to blunt the transient (hours to a couple of days) fevers and chills that come with the infusion of this class of drugs, and sits there for about an hour while Provenge is infused via an IV. Eduardo goes home the same day.
Intellivenge tracks all the steps of this. Once successful infusion is completed, Intellivenge repeats the process two more times. (In fact, Intellivenge schedules all three manufacturing runs at the same time but I’ve simplified it here to make the process more clear.)
There are weaknesses in the system, of course. The biggest is making sure the patient shows up on time to the appointments. Weather and other reasons can interfere with this. Intellivenge is designed to pretty well pester the patient and doctor to make sure this issue is minimized. I personally know from talking to many prostate cancer patients and patient groups that they're very motivated to obtain a drug with Provenge’s profile -- big efficacy compared to other available treatments and almost no side effects. My view is this sort of hiccup will be rare.
The Financial Side
I noted above where the doctor gets charged in the process. This is important because the doctor is responsible for making sure he/she gets reimbursed. Medicare will be responsible for 70-80% of patient reimbursement, and they can be notoriously slow. That delay combined with Provenge’s $50,000-$100,000 per patient price tag (my estimate, not Dendreon’s) means there could be a significant cash flow crunch at the PAD.
Dendreon has thought of that. They have a contract with a third party to create a “flooring plan” for Provenge. The third party will issue a float loan to the PAD covering the time between when payment is due to Dendreon (upon delivery of the dose to their offices) and when the PAD is reimbursed by Medicare or a private insurer. Use of the flooring plan is optional, of course. Some PADs have enough cash reserves so they won’t need the cash flow help.
A common question is what the cost of goods will look like for Provenge given the process outlined above. I’ve shared some guesses with our clients and have seen guesses from others. Dendreon has indicated they’ll give some guidance after approval. Some people are obsessing about this, claiming even small hiccups in the manufacturing process will render that patient unprofitable. After 10 years covering the company, I can state unequivocally that Dendreon isn’t that stupid. The margins will be more than sufficient. Of note, Dendreon CFO Greg Schiffman said the company’s most recent equity financing would be its last. They will be able to generate more than enough cash from the sale of Provenge to handle operations and R&D on their pipeline going forward. That wouldn’t be possible with very narrow margins.
Dendreon is one of the more fascinating companies I’ve covered in the 10 years our firm has been biotech specific. Provenge is an honest-to-goodness paradigm shift in the way we treat cancer. Good news from their IMPACT trial helped the entire biotech sector bottom early last year. Approval and strong sales should help the sector in 2010 and 2011.
Intellivenge is quite the accomplishment. It’s been personally fun watching the management team work out all the potential kinks and twists, attempting to account for every variable. While they're smart enough to know five nines perfection is impossible with this much human interaction, the combination of Intellivenge, the call center, and significantly motivated patients will minimize surprises considerably.
Our firm has projected Provenge will achieve over a billion in sales in the over the next few years. We’ve shared with our clients specific sales projections for 2010 and 2011. Managing demand will be Dendreon’s biggest challenge until full manufacturing capacity is online in mid-2011.
That’s a nice problem to have.
The information on this website solely reflects the analysis of or opinion about the performance of securities and financial markets by the writers whose articles appear on the site. The views expressed by the writers are not necessarily the views of Minyanville Media, Inc. or members of its management. Nothing contained on the website is intended to constitute a recommendation or advice addressed to an individual investor or category of investors to purchase, sell or hold any security, or to take any action with respect to the prospective movement of the securities markets or to solicit the purchase or sale of any security. Any investment decisions must be made by the reader either individually or in consultation with his or her investment professional. Minyanville writers and staff may trade or hold positions in securities that are discussed in articles appearing on the website. Writers of articles are required to disclose whether they have a position in any stock or fund discussed in an article, but are not permitted to disclose the size or direction of the position. Nothing on this website is intended to solicit business of any kind for a writer's business or fund. Minyanville management and staff as well as contributing writers will not respond to emails or other communications requesting investment advice.
Copyright 2011 Minyanville Media, Inc. All Rights Reserved.