http://seekingalpha.com/article/3651256-victory-in-the-car-t-race-may-be-pyrrhic-juno-the-biggest-loser-cellectis-too-late-for-the-spoils?page=2
Summary
"T cell manufacturing is robust and dependable...A handful of academic centers have successfully established manufacturing procedures that have proven to be reproducible and dependable" (Sadelain, 2015).
"The field ... has a dearth of data as to whether centralized manufacturing of patient-derived CAR T cells is broadly feasible and does not result in ruinous costs..." (Cooper, 2015).
"Amateurs talk about tactics, but professionals study logistics" applies to military ops but also to clinical ops being transitioned to commercial, centralized manufacturing.
CEOs who think they can take this brilliant science commercial without a road map through the daunting logistics prove themselves amateurs.
Avoid them all until you at least see a road map, especially the biotechs with largest market cap. Consider small-cap contract manufacturers who demonstrate expertise in logistics.
A victory is pyrrhic when even the victor counts the cost too high, and that is how the crowded race for autologous CAR-T cells could end.
Genetic engineering of CARs into T cells is one of the most complex processes known to modern medicine. Central as opposed to point-of-care manufacturing of CAR-T cells for autologous transfusion adds logistical complexity and treatment risk to an intense therapy that already carries great risk to a vulnerable patient group, those with cancer refractory to conventional therapies.
Experts in commercial scale manufacturing warn against the scaling up of immune cell production without a transition period of close collaboration between manufacturer and the academic facilities in which patient-specific, genetically engineered T cell therapies originate: the science behind CAR-T is brilliant but scaling up central manufacturing requires a vastly different skill set (see
Preti, 2015 and
O'Donnell, 2015).
No one should know more about failed cell manufacturing models than Juno (NASDAQ:
JUNO) CEO Hans Bishop who was EVP and COO at Dendreon (
OTCPK:DNDNQ) from January 2010 to September 2011. Dendreon attempted to manufacture its own cellular therapeutic. DNDNQ entered Chapter 11 in November 2014.
Dendreon's PROVENGE® sipuleucel-T is a patient-specific but centrally manufactured immunocellular therapy for autologous transfusion in prostate cancer. Despite a favorable impact on overall survival (Kantoff, 2010), the medical community did not think the benefits of Sipuleucel-T justified the $93,000 price tag, sales were disappointing, production costs high - a Dendreon
website video shows why: logistics are a potential nightmare.
Harvesting cells from a patient, shipping them to a manufacturer for immunological manipulation, and sending them back for transfusion into the same (we hope) patient is not only risky business, it is risky medicine: Dendreon had difficulty maintaining quality control amidst the complexities of manufacture and distribution. I am not saying that cancer therapy with CAR-T will be as unimpressive as Provenge has been. T cell therapies are poised to become part of the standard-of-care treatments for patients with cancer (Wang and Rivière, 2015). But manufacturing genetically engineered CAR-T cells is longer (2 weeks) and more complex than for Provenge (not genetically modified and takes 3 days) by an order of magnitude.
In a white paper, titled "Manufacturing Cell Therapies: Development Strategies for Commercial Vision", authors Richard Grant and Brian Hampson, VP of Manufacturing Development and Engineering for
Progenitor Cell Therapy ((NASDAQ:
CLBS)), offer a fairly detailed road map which I have summarized:
- Client (e.g. Juno) creates a Quality Target Product Profile for future labeling that follows FDA guidelines,
- Long-term collaboration begins with manufacturing system developer (e.g. PCT/CLBS) visiting client to observe processes currently in use
- Client formally communicates to manufacturer expectations for numbers of patients to be treated, costs of products, handling and shipping of products, and other requirements
- client gains a clearer strategic vision to communicate to investors and others
- CMO (e.g. Invetech/CLBS) builds prototype for small-scale production, produces cells for quality testing by client
I searched JUNO's latest 10K, 10Q and earnings conference calls for evidence that Hans Bishop had some learned some important lessons from the DNDNQ failure, and communicated to investors (as in step 4 above) his road map for success. This is what I do see:
"We've transitioned from an academic process to a commercially scalable one."
How?
"The company entered into a lease for an approximately 68,000 square foot manufacturing facility in Bothell, Washington, which lease commenced in March 2015."
Already?
"We plan to complement the use of one or more CMOs by establishing our own cGMP manufacturing facility to be brought on-line after the first CMO. As described...we have entered into a ten-year lease for a facility that we plan to remodel to support our clinical and commercial manufacturing activities."
Headlong into uncharted waters is what that sounds like to me. Meanwhile, Novartis has already moved into the NJ cell manufacturing plant that Mr. Bishop
equipped, and his successor at DNDNQ had to
liquidate.
Novartis is too big a company for me to study and present, but I took note of its collaboration with U Penn when the CART19 story made big news in NEJM (Porter, 2011). NVS surely has the liquidity to stay in this CAR-T race longer than JUNO could stay solvent without the aid of Celgene (NASDAQ:
CELG). And what an advantage having a ready-made (courtesy of Hans Bishop's Dendreon) cell manufacturing plant when all JUNO has to show its investors is a lease agreement!
David Chang, Kite's (NASDAQ:
KITE) chief medical officer, told a
Reuters reporter that Kite will minimize overhead and manufacturing costs by building individual modules that can each handle a patient's cells and then will then "build out" with more factory modules as demand increases. That sounds like a good idea, and I think Kite's plan for a commercial manufacturing facility adjacent to Los Angeles International Airport also shows they are logistically minded. Meanwhile, KITE will leave responsibility for cell manufacture to
Progenitor Cell Therapy, a subsidiary of Caladrius Biosciences. CLBS bioprocesses and manufactures but does not genetically engineer cells. CLBS operating expenses exclusive of R&D rose faster than revenues in FY 2013 and 2014, and its own cell therapy programs are not doing well. President of this company, Dr. Robert Preti, has posted an article in DDNews, September 2015, provocatively entitled "
Building a problem or a solution?" Dr. Preti articulates concerns about the state of his industry and the direction it must go. KITE's dependence on this tiny company with only $21M in PP&E is of some concern - CLBS will need capital to build capacity, and that will be more difficult with its stock price way down this year. Perhaps CLBS will be acquired.
There are some sub-$1B market cap companies that are engineering features into T cells to make them safer and more effective, but for whom logistics is not yet a critical concern.
Bellicum (NASDAQ:
BLCM) has a variety of T lymphocyte and other immune cell platforms. Its most advanced product in phases 1 and 2 is BPX-501, an allogeneic, partially matched T cell removed from donor bone marrow prior to transplantation, engineered with a suicide switch, then used to reconstitute the immune system of the recipient of the transplant. Genetically adding suicide switches (Stasi, 2011; Zhou, 2014) allow for these cells to be shut down if they attack the patients own healthy tissues. Bellicum does research in partnership with scientists at Baylor. BPX-201 is a preclinical cancer vaccine that uses a prostate cancer patient's own dendritic white cells for autologous infusion after genetic engineering of a suicide switch and expression of a prostate cancer antigen that will stimulate and immune attack on the cancer. BPX-601 and -401 are preclinical CAR-T cells with suicide switches. Both are for autologous transfusion, so centralized manufacturing for wide distribution will face the same logistical issues.
Adaptimmune (NASDAQ:
ADAP) plans to build its manufacturing plant in
The Navy Yard in Philadelphia for use in 2016. Ships are known to move slowly, so logistics was probably not a consideration in selecting that location. The ADAP platform is entirely autologous and not off-the-shelf. ADAP has 8 phase 1 or 1/2 trials registered at ClincialTrials.gov, all using T cells with genetically engineered TCR, not CARs. One of these trials which uses anti-NY-ESO-1 TCR for synovial sarcoma is in collaboration NCI. GlaxoSmithKline (NYSE:
GSK) is not named as co-sponsor on any of the registered trials but has options on the T cells with TCR engineered against NY-ESO. ADAP form 10-K indicates (page 56) that current CMO is Progenitor Cell Therapy.
Pity that Cellectis (NASDAQ:
CLLS) (
OTCPK:CMVLF), the only company developing a universal, off-the shelf CAR T product that would be logistically simpler to manufacture and distribute, has registered nary a single clinical T cell trial among the numerous already registered at ClinicalTrials.gov (see Table below). Not only that, but in reviewing a large amount of the literature on the subject, I came across not a single paper authored by Cellectis scientists, and no mention of them by peers in the field. When I first discovered Cellectis in a review of the 2014 ASH abstracts, I thought I had discovered a gem - and it may indeed be a gem. But Cellectis promised a clinical trial in 2015 which as of November has not started or been further discussed. Lead product is another anti-CD19 CAR, but the race against CD19-bearing B-cell malignancies is getting crowded, and Cellectis may have difficulty finding a market by the time its CRO has gotten UCART19 through trials and approved. The "barrier to entry" into the universal donor CAR-T race is not high: it requires that the lymphocytes native TCR be knocked out, and that technology is currently being widely adopted by others.
Conclusions.
- Central bioprocessing and manufacturing of genetically engineered T cells for autologous transfusion is an unproven business model. Those who try it should learn from Dendreon's failure with a much simpler, genetically un-modified cell therapy product.
- Even industry experts admit (Kaiser, 2015):
"It is conceivable to see a future where gene-modified T cells are manufactured at the point-of-care in a facility in close proximity, associated with, or at the hospital."
...that's quite an understatement
- The cost of manufacturing an off-the-shelf, allogeneic rather than autologous, cancer-fighting T cell will be less ruinous.
- Beyond the economic fundamentals, identification the most important and safe molecular targets, and the ability to edit T cell genes translate competitive edge (June, 2015).
Recommendations:
- $5B JUNO is a STRONG SELL
- $3B KITE is a SELL
- $0.9B Cellectis is very late to enter the race particularly for a CD19 CAR-T, but has a platform for allogeneic use. Speculative HOLD.
- $0.8B market cap Adaptimmune, new to the market in May, was noticed by another SA author in September and is worth following.
- $0.3B BLCM is worth watching
- $68 Million market microcap CLBS is also worth watching. CLBS does have the skill set for manufacturing, and its stock will soar if BLCM or ADAP go commercial before they are fully cGMP-capable.
