2:53 PM
Apr 13, 2018
 |  BioCentury  |  Strategy

Self-driving CARs

Why Pfizer and other large companies think CAR T technology needs its independence

Editor's Note: This article was updated on Sep 14, 2018 at 6:20 PM PDT

Novartis AG, Gilead Sciences Inc. and now Pfizer Inc. have taken different approaches to managing their CAR T cell assets, but the three companies appear to be converging on the notion that developing the technology requires a level of autonomy that is unavailable within the traditional biopharma operating model.

Each has chosen to house its CAR T cell pipelines in an entity that operates independently of the rest of its pharmaceutical operations.

Novartis, an early adopter of the modality, also was the first to decide it needed to be managed differently from small molecules or biologics. The pharma set up a dedicated cell and gene therapy unit in 2014, just as Kymriah tisagenlecleucel was about to enter a pivotal Phase II study. The pharma later folded the unit into its Novartis oncology business, which operates separately from the larger pharmaceuticals unit.

When Gilead acquired Kite Pharma Inc. last October, it took a similar decision, allowing the biotech to operate as an independent business unit.

Pfizer decided a greater degree of separation was necessary for the off-the-shelf CAR T cell therapies it was developing using TALEN gene editing technology from Cellectis S.A. The pharma announced April 3 that it would spin out the portfolio into newco Allogene Therapeutics Inc.

Both Novartis’ decision to fold cell and gene therapy into its oncology unit, and Pfizer’s decision to spin its CAR T portfolio out led analysts and investors to wonder aloud whether the pharmas were retreating from the technology.

But it appears that both moves reflect the unique needs of cell therapy development, particularly with respect to the interdependence of manufacturing, R&D and regulatory groups.

“It’s not about a lack of belief that cell-based therapies are going to be transformational in certain cancer types,” Pfizer’s Robert Abraham told BioCentury. “We needed to get this into the hands of a group that we had extremely high confidence in to optimize the use of the platform and who would have CAR T cells as a single mission objective.” Abraham is SVP and head of oncology R&D group at Pfizer.

Allogene launched with Kite’s former management team, $300 million in funding and 16 CAR T cell candidates. David Chang, former EVP and CMO of Kite, is president and CEO of Allogene. Former Kite Chairman, President and CEO Arie Belldegrun serves as executive chair.

“It’s not about a lack of belief that cell-based therapies are going to be transformational in certain cancer types.”

Robert Abraham, Pfizer

Two programs are disclosed: the Phase I UCART19 program and a preclinical CAR T targeting tumor necrosis factor (TNF) receptor superfamily member 17 (BCMA; TNFRSF17; CD269). The remaining CAR Ts are in preclinical development to treat hematologic malignancies and solid tumors.

Pfizer will have a 25% stake in the company. Pfizer did not disclose whether it retained any clawback rights or rights of first negotiation to any of Allogene’s 16 programs as monotherapies or as part of combinations with Pfizer agents.

Pfizer’s decision was based in part on a prioritization exercise that concluded the company could not devote the resources to getting CAR Ts through the clinic at the same time it managed its oncology pipeline of small molecules and biologics.

UCART19, a CD19-directed allogeneic CAR T cell, is due to enter pivotal studies next year -- close to the same inflection point where Novartis decided to give more autonomy to its cell therapy unit.

A common theme across all three companies’ decisions to separate their CAR T operations from their pharmaceutical units is that CAR T technology needs an organization that can adapt quickly to emerging science, and in which manufacturing and R&D are far more tightly linked than in typical small molecule or biologics development.

Even with Kymriah and Gilead’s Yescarta axicabtagene ciloleucel now on the market, there is still much to sort out with this new modality -- including how to crack solid tumors, reduce sometimes lethal side effects and improve the cumbersome manufacturing process.

Building the wall 

When the first CAR Ts entered the clinic in 2010, they were still the domain of academia and research institutions such as the University of Pennsylvania and the National Cancer Institute (NCI).

The modality started to attract industry attention in 2012 when Novartis licensed the CD19-directed CAR T now known as Kymriah from UPenn. It quickly concluded these modalities required a different touch than small molecules or antibodies.

“When we brought the CAR T in from UPenn and then tried to develop it in a traditional structure, I realized it wouldn’t work,” said David Epstein, who was CEO and division head of pharmaceuticals at Novartis at the time. Epstein is now chairman of cell therapy company Rubius Therapeutics Inc. and executive partner at Flagship Pioneering.

Epstein said the pharma’s operating structure was bogging the cell therapy programs down.

Under the traditional model, manufacturing and drug development are delinked, with limited interaction. But the cell therapy manufacturing process, particularly in its early stages, can evolve on a patient-by-patient basis, requiring close interaction between manufacturing and clinical development.

“As the data come in and you see how the cells expand, you are making changes to the manufacturing process on the fly. But manufacturing at big companies isn’t set up this way, they are more disconnected,” Epstein said.

After Epstein recruited Usman Azam to lead the cell and gene therapy unit, “it became very accelerated after that point,” he said.

The pivotal Phase II study began in 2015. FDA approved Kymriah in August 2017 to treat relapsed/refractory pediatric B cell acute lymphoblastic leukemia (ALL). An sBLA is under FDA review to treat relapsed/refractory diffuse large B cell lymphoma (DLBCL).

Novartis has three additional CAR T products in Phase I testing.

After Epstein departed in 2016, Novartis split is pharmaceuticals R&D business into two -- oncology and pharmaceuticals. At the same time, the cell and gene therapy was moved into the Novartis Oncology business.

Novartis made the move to allow “the oncology business to capitalize opportunities, confront challenges and accelerate growth,” the company told BioCentury at the time.

Spokesperson Eric Althoff said Novartis has integrated CAR T manufacturing into its oncology business and added scale.

It also continues to build on the technology through early stage collaborations via its Novartis Institutes for BioMedical Research, including its partnership with gene editing company Intellia Therapeutics Inc., Althoff said.

Flying free 

Gilead took a similar approach with Kite, which has become the big biotech’s dedicated cell therapy unit.

The company declined to provide an interview. It has not described the unit’s operations further, or said why it prefers to maintain it as a distinct unit.

The unit has, however, continued to do deals. For example, in December Kite and Gilead together acquired the 87.8% of Cell Design Labs Inc. that Kite did not already own for $175 million up front and up to $322 million in milestones. Cell Design’s synthetic CAR T technology includes a Throttle Switch platform to control the potency and duration of CAR T activity.

In February, the Kite unit partnered with Sangamo Therapeutics Inc. to use zinc finger nuclease (ZFN) and adeno-associated virus (AAV) technology to develop next-generation autologous and allogeneic cell therapies to treat cancer.

With Chang and Belldegrun leaving Gilead, EVP of Oncology Therapeutics Alessandro Riva will assume leadership of the CAR T pipeline.

Riva was head of global oncology drug development at Novartis until January 2017, when he joined Gilead. He oversaw development of Kymriah for the Swiss pharma.

At the time of the Kite acquisition, Epstein told BioCentury that Riva understood well “the intimate connection of every decision between the clinical development, manufacturing and regulatory teams” that is necessary for cell therapy technologies.

Need for Allogene 

After nearly four years of investment in immuno-oncology, Pfizer came to the realization that a stand-alone business would be the best vehicle for its CAR Ts.

The pharma entered the field in 2014 via its partnership with Cellectis. The collaboration gave Pfizer rights to develop and commercialize CAR T therapies for 15 cancer targets.

Celgene Corp. and Novartis already were in the clinic with autologous cell therapy candidates, but Pfizer was attracted to the off-the-shelf nature of an allogeneic product, and to Cellectis’ gene editing technology to remove T cell receptors to reduce the immunogenicity of donor cells.

It licensed U.S. rights to co-develop and commercialize UCART19 from Servier in 2015. The allogeneic CAR T therapy, which also was developed by Cellectis, entered the clinic in 2016 to treat pediatric and adult ALL.

“We felt that putting these into a company with a singular focus should be able to drive this to clinic quicker than we could.”

Robert Abraham, Pfizer

At the time of the Cellectis deal, Pfizer had just one immunotherapy in the clinic. It now has nine immunotherapy candidates and more than 25 ongoing combination studies of its anti-PD-L1 therapy Bavencio avelumab.

Faced with the need to ramp up manufacturing and get its preclinical CAR Ts into the clinic, the pharma had to make some choices.

“We have an emerging immuno-oncology portfolio but we also have limited funds and everything competes,” said Abraham. “We felt that putting these into a company with a singular focus should be able to drive this to clinic quicker than we could.”

Pfizer will have two seats on Allogene’s board along with its equity stake in the company. It also has an 8% stake in Cellectis.

“We are committed to investing in immuno-oncology but for the time being with cell-based therapies, we will leave those to others. We may jump in again later,” Abraham said.

Loncar Fund’s Brad Loncar thinks that Pfizer made the right decision.

“These therapies are best developed in a totally dedicated, entrepreneurial-type environment,” he said. “A company like this can be totally dedicated to riding the ups and downs of the technology and getting it across the finish line.”

Taking talent 

Pfizer said it had multiple offers, but was attracted to the talent and experience Chang and Belldegrun brought.

“They are obviously steeped in the needs of the technology. They have a network of academic investigators who are engineering the next generation of cell therapies, and an interesting set of assets that have now been moved over to them,” said Abraham.

Chang told BioCentury he was approached about the opportunity in November by a banker from Centerview. While still at Kite, he’d been following the allogeneic field as a potentially faster and more standardized method to manufacture the therapies.

“We were keenly aware that there were inherent limitations with autologous cell therapy,” Chang said. “For almost four years, we’ve been looking into this from a scientific and technology perspective and we knew the space really well, including the Cellectis platform.”

Pfizer has not published much on the work it has done in CAR T, and Chang declined to describe the data Allogene reviewed in its due diligence.

He did say, “We were amazed at a lot of improvement and different targets that they had developed.”

He added that Pfizer’s data showed proof of concept, but declined to say whether or not the data that led to that conclusion had been published.

The clinical data Pfizer and Servier presented for UCART19 were underwhelming.

In December 2017, Pfizer and Servier presented Phase I data at the American Society of Hematology (ASH) meeting showing molecular remission in pediatric and adult B cell positive ALL, but half of the patients that achieved remission relapsed following subsequent stem cell transplant.

All five pediatric patients treated with UCART19 had a molecular remission at day 28, defined as negative minimal residual disease (MRD). Two relapsed following transplant.

“In our close look at what they’d done, we were amazed at a lot of improvement and different targets that they had developed.”

David Chang, Allogene

Five of seven adult patients in the Phase I CALM study treated with UCART19 had complete remission defined as MRD-negative at day 28. Three later relapsed.

There aren’t comparable data from autologous products. In the pivotal pediatric ALL study of Kymriah, 52 of 63 (82.5%) patients had a complete remission within three months. Of these, seven went on to receive stem cell transplant. Outcomes for those patients have not been reported. Median overall survival was 16.7 months.

The only other program Pfizer has presented is CART-BCMA.

At the 2016 ASH meeting, Pfizer and Cellectis presented preclinical data showing the therapy killed BCMA-expressing multiple myeloma (MM) cell lines. BCMA is found on the surface of nearly all MM cells.

The presentation also highlighted features of CART-BCMA designed to improve its performance and avoid rejection of cell therapy in vitro. The product contains a rituximab-activated switch to shut down the CAR T when necessary, along with a lymphodepletion resistance mechanism engineered in via CD52 knock out.

Before administering other CAR T therapies, patients receive the anti-CD52 mAb Campath alemtuzumab for lymphodepletion to prevent or reduce graft-versus-host disease (GvHD). But Campath could reduce the expansion and efficacy of CAR T cells.

Knocking out the CD52 domain of the T cell protects CART-BCMA from Campath-induced toxicity. Data presented at ASH showed that these changes did not reduce the CAR T’s antitumor activity.

From auto to allo 

Allogene intends to improve its manufacturing process to juice the efficacy and safety of the therapies.

Unlike autologous cell therapies, allogeneic cells that have T cell receptors can trigger deadly immune reactions such as GvHD. TALEN removes T cell receptors from donated cells, but Chang said it’s typically about 80-90% effective.

“The remaining T cell receptors have to be purified during the manufacturing process. Pfizer’s scientists have done a lot of that to make it significantly better, but we think it’s an area we can also improve on,” he said.

He also said Allogene would optimize the lymphodepletion technology and “make the manufacturing process more efficient.” He declined to provide details.

Cellectis expects Allogene will be able to speed the allogeneic products through the clinic.

“Their execution with Yescarta was excellent. They already have shown the ability to set up a manufacturing process with a low failure rate, which skyrockets the chances of these products getting into registration as fast as possible,” said Chairman and CEO Andre Choulika.

Manufacturing failures in autologous cell therapy include failure of the T cells to take up modifications during the transduction process, or to expand sufficiently. According to FDA review documents, among the 110 patients who underwent leukapheresis in Kite’s pivotal Phase II study of Yescarta to treat non-Hodgkin’s lymphoma (NHL), there was one manufacturing failure.

For Cellectis, speeding the path through the clinic would mean getting its milestone payments sooner. The biotech is eligible for up to $2.8 billion in milestones on the products, plus tiered royalties.

Companies and Institutions Mentioned 

Allogene Therapeutics Inc., South San Francisco, Calif.

American Society of Hematology (ASH), Washington, D.C.

Cellectis S.A. (NASDAQ:CLLS; Euronext:ALCLS), Paris, France

Gilead Sciences Inc. (NASDAQ:GILD), Foster City, Calif.

Novartis AG (NYSE:NVS; SIX:NOVN), Basel, Switzerland

Pfizer Inc. (NYSE:PFE), New York, N.Y.

Rubius Therapeutics Inc., Cambridge, Mass.

Sangamo Therapeutics Inc. (NASDAQ:SGMO), Richmond, Calif.

Servier, Suresnes, France

U.S. Food and Drug Administration (FDA), Silver Spring, Md.


McCallister, E. “Some of the parts.” BioCentury (2017)

McCallister, E. “Two heads are better.” BioCentury (2016)

McCallister, E. “Pfizer moves up.” BioCentury (2014)

McCallister, E. “Three-patient POC.” BioCentury (2012)

Shaffer, C. “Pfizer picks its CAR T.” BioCentury (2014)

User Sign in

Trial Subscription

Get a 4-week free trial subscription to BioCentury

Article Purchase

Purchase this article for limited one-time distribution and website posting

$750 USD