If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



Making natural killers stronger against cancer

Engineered NK cells are part of the next generation of off-the-shelf CAR therapies

by Megha Satyanarayana
July 28, 2019 | A version of this story appeared in Volume 97, Issue 30


A doctor pulling a vial of cells off a shelf.
Credit: Will Ludwig/C&EN

Cancer immunotherapy has traditionally been the realm of the T cell—an abundant and malleable killer that has been made deadlier through a class of antibody drugs called checkpoint inhibitors and a type of engineered cell known as CAR-Ts.

But T cells aren’t the only immune cells that can sniff out cancers. With the enormous success rates of CAR-Ts, the race is on to modify other types of immune cells to feature CARs, or chimeric antigen receptors, proteins that are built to recognize cancerous cells. And nowhere is the buzz stronger than around a cell appropriately called the natural killer.

Natural killer cells, or NK cells, are part of the next wave of cell-based cancer treatments that promise to treat more people for less money than currently available CAR-T therapies, which are currently tailor-made for each cancer patient. Their “off-the-shelf” versatility, plus the variety of CARs and other proteins that can be introduced into NK cells, makes them a tantalizing prospect to expand the reach of immunotherapy. More than a dozen clinical trials are underway worldwide to test engineered NK cells on a variety of blood cancers and solid tumors, and early results seem to suggest the treatments are safe. Now many wait to see if CAR-NKs will be as effective as, if not more effective than, their T-cell cousins.

“NK cells have an underappreciated potential as a cancer therapy. They’re really multifaceted killers,” says Dan Shoemaker, chief scientific officer of Fate Therapeutics, a biotech with two NK cell–based immunotherapy products in early clinical trials. Human data are still being amassed, but Fate and others suggest that NK cells “might really give T cells a run for their money.”

When it comes to cell engineering, T cells are perhaps easier cells to work with, says Katy Rezvani, a researcher leading CAR-NK clinical trials at MD Anderson Cancer Center. Not only are T cells more abundant than NK cells, but CARs were designed to mimic the action of T-cell receptors that recognize foreign protein pieces, or antigens, on the surfaces of other cells.

To make T cells more targeted cancer killers, these receptors have been rejiggered to create CAR-T therapies. For example, Novartis’s Kymriah, the first US Food and Drug Administration–approved CAR-T treatment, is made of parts of four proteins fused together: an antigen receptor that appears on the surface of the engineered T cell, a domain that goes through the cell membrane of the T cell, and two domains that extend into the cell. When the receptor is bound to its target, those domains stimulate the T cell to kill the cancer cell and also tell the T cells to grow, divide, and persist long enough to keep killing. For Kymriah, the antigen receptor recognizes a protein called CD19, which is expressed highly in some blood cancers.

A natural killer cell attacking a cancer cell.
Credit: Science Source
Natural killer cells (yellow/blue) can find and kill cancer cells (red), as shown in this colorized photo. They are part of the next wave of engineered cell therapies.

But NK cells’ cancer-killing prowess is also well known. One of the first things scientists noted about them is that they killed cancer cells differently from T cells. NK cells are rovers—they can sniff out infected or cancerously transformed cells in a way that is less stringent than T cells. Once NK cells latch on, they release proteins that destroy their sickly hostages.

“Natural killer cells are the most cytotoxic immune effectors that we have,” Rezvani says.

That killing ability spurred early trials of nonengineered NK cells as a possible cell-based cancer treatment. In the early 2000s, people with cancer were given millions of donor NK cells. But Rezvani says that the therapy, while relatively safe, was not particularly effective (Blood 2005, DOI: 10.1182/blood-2004-07-2974). Getting NK cells to take up CARs and other genetic constructs to amplify their natural talents and make them more powerful cancer killers has required some finesse.

But NK cells have several distinct advantages over T cells. First, NK cells don’t last as long as T cells, meaning the likelihood of dangerous side effects is low, Rezvani says. CAR T cells are designed to persist in the body, and one side effect is what’s known as a cytokine storm, a rapid release of immune proteins that can overwhelm the body and damage organs. Oncologists have learned to control this, but in the early days of CAR-Ts, cytokine storms killed several people with cancer.

CAR NK cells’ short life span may circumvent such damaging cytokine release—early clinical trials seem to support this—but it also means that people getting CAR-NK treatment will likely need multiple infusions, compared with a single dose of CAR-Ts, Fate’s Shoemaker says.

NK cells are also more versatile than T cells. Currently approved CAR T cells can go back only into the person they came from. T cells can tell the difference between one person and the next—the body of a person with cancer will reject T cells from another. NK cells do not have this restriction. In fact, Rezvani says there’s evidence that NK cells work better if they come from a donor. Researchers are getting their NK cells from different sources: Rezvani’s lab uses cord blood, Fate uses stem cells that it turns into NK cells, and Nkarta, another biotech specializing in NK cell therapies, uses NK cells from donors’ blood.

Otherwise, the protocols required to engineer a CAR onto NK cells are similar to those needed to do the same to T cells. First, cells are isolated from starting materials, be it blood or cell culture. Scientists then use viruses to insert a gene encoding for the CAR that when expressed will help the cells more readily recognize and kill cancer. The next step is to isolate the cells with the CAR and make vast numbers of them. But while CAR-Ts then must go back into the patient whose cells were engineered, CAR NK cells can be frozen to be used in multiple people.

Some of the most advanced CAR NK-cell therapies express receptors that target CD19, much the same as Kymriah and the other approved CAR-T treatment, Gilead Sciences’ Yescarta. But researchers are also adding extra proteins to their CAR-NKs in the hopes of boosting their power and ability to persist long enough to be effective.

In one of the first CAR-NK trials, a clinical team in China studied the safety of CAR-NKs in three people with a blood cancer called acute myeloid leukemia.The researchers isolated NK cells from a line treated to prevent the cells from dividing once administered to a patient. Jianhua Yu, who oversaw the trial while at Ohio State University, was looking for whether NK cells might seed the rapid cytokine release that once troubled CAR-Ts. The treatment appeared to be safe.

Fate’s most complex NK product, FT596, has a CD19 CAR but also a CD16 receptor and an IL-15 receptor. The CD16 receptor serves as a backup if the tumor has wiped CD19 from its surface, and the IL-15 receptor helps NK cells grow and persist.

Rezvani’s trial at MD Anderson Cancer Center involves NK cells with a CD19 CAR and an IL-15 receptor. She expects to release her team’s full findings soon but says the treatment appears safe.

Nkarta is also working on a CD19 CAR, which the company hopes to put into human studies by the end of 2020. But the firm’s lead product, NK101, is an NK cell with a CAR based on a receptor called NKG2D, which recognizes several proteins on the surface of cancer cells. Nadir Mahmood, Nkarta’s head of corporate development, says that by overexpressing the NKG2D CAR, the firm hopes to put its cells into cancer-scouting overdrive. NK101 should be in clinical trials by early 2020, and Nkarta wants to test it in people with acute myeloid leukemia as well as those with solid liver cancers.

Should CAR-NK therapies reach the market, researchers believe another benefit will be the cost: CAR-T therapy currently carries a list price of around $350,000, but the off-the-shelf nature of CAR-NKs could drive the cost down to the order of thousands of dollars per dose, Shoemaker says.

“It’s only going to get cheaper as we get better at this,” he says, referring to the potential for large batches of cells to treat multiple people. “It can’t be hundreds of thousands of dollars per dose.”

Rezvani agrees and marvels at how far the field has come since her days as a graduate student in the late 1990s, when she says people thought of immunotherapy as “voodoo medicine.”

Even as the field slogs through challenges, scientists are optimistic about engineered NK cells—they might treat not only cancer but infectious diseases and other illnesses as well. “We live in a very exciting time. The next decade or so is going to be a true revolution in the field of cell therapy,” Rezvani says.


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.