The gargantuan, hanger-like bay of the San Diego conference center was a whirlwind of activity earlier this month, as thousands of the world’s cancer researchers scientists poured in for the annual meeting of the American Association for Cancer Research (AACR), held April 5-9, 2014.
Hundreds of symposia were held simultaneously and back-to-back over the course of the five-day event. Topics spanned the entire field of cancer research, from molecular studies of cell cycle proteins to the role of vegetable consumption in cancer prevention. The clear hot topic of this year’s conference, however, was cancer immunotherapy. More than 30 talks and dozens of poster sessions were devoted to the treatment modality.
Parade of PD-1
A particular source of buzz was the second checkpoint pathway to be discovered, PD-1. Like its more famous older sibling, CTLA-4, PD-1 is a protein found on immune cells that acts as a brake, or checkpoint. Antibodies generated against PD-1 (anti-PD-1) or its binding partner (anti-PD-L1) block the pathway's activation, in effect releasing the brakes on the immune response.
PD-1, which stands for programmed death receptor-1, was identified by Tasuku Honjo, M.D., Ph.D., of Kyoto University School of Medicine. Its function as a negative regulator of T cell function that could be exploited in cancer treatment was independently discovered by Lieping Chen, M.D., Ph.D., of Yale University School of Medicine.
PD-1 has generated intense excitement among cancer biologists because of the promising responses shown in clinical trials using anti-PD-1 as both monotherapy and in combination with other drugs, including ipilimumab (anti-CTLA-4).
Among the most exciting findings of these studies is that a significant proportion of patients with non-small cell lung cancer (NSCLC) respond dramatically to anti-PD-1 treatment. Lung cancer has long been thought to be a cancer that would not be susceptible to immune-based treatments, but these results clearly show that view to be mistaken.
In fact, some researchers are even suggesting that lung cancers use the PD-1 pathway to escape immune destruction in the first place. The most commonly mutated oncogene in NSCLC is the epidermal growth factor receptor (EGFR). EGFR is a protein found on many cells in the body that provides a signal for growth. In cancer, this protein becomes hyper-activated, leading to abnormal growth.
CRI associate director Glenn Dranoff, M.D., and his colleagues at the Dana-Farber Cancer Institute in Boston presented intriguing results from their lab showing that mutations in EGFR lead to increased activation of the PD-1 pathway in the tumor environment. What this suggests is that lung cancers are able to take hold in the lungs by disarming T cells and escaping the immune system’s control.
PD-1 was clearly the most talked-about molecule at the conference, prompting one scientist to confess, “I feel like an exhausted T cell.” But all of the current approaches of cancer immunotherapy were well represented.
T Cell Tune-Up
Some of the most dramatic images of the conference came from researchers affiliated with Steven Rosenberg’s lab at the National Cancer Institute. For many years, Rosenberg’s group has been experimenting with a form of immunotherapy called adoptive T cell transfer. T cells that specifically recognize cancer are isolated from the patient, stimulated with IL-2 (a growth factor for T cells), expanded into billions of copies, and then replaced into the patient. When it works, this approach can lead to near-miraculous recoveries. Grapefruit-size tumors protruding from necks and limbs were shown to wither away to nothing.
Yet there were also many questions raised at the meeting. Nicholas Restifo, M.D., of the NCI, asked: “Why do only some people respond to CTL treatment?” This is a common refrain among cancer immunologists, who to this day just don’t have a good way to predict which patients will respond to immunotherapies and which will not. The biomarkers that might provide an explanation have so far proved elusive, though researchers are making in-roads.
In the case of adoptive T cell transfer, part of the answer may lie in the particular cytotoxic lymphocytes (CTLs) that are transferred to a patient. Researchers know that patient prognosis generally correlates with the length of time the tumor-specific CTLs hang around in the body after infusion. The question is: why do the cells hang around in some patients and not others?
Restifo and colleagues have shown that mature, highly reactive T cells tend not to persist very long in the body. Younger, less differentiated T cells, on the other hand, hang around for much longer. The really interesting thing, though, is that an entire population of young T cells can be “corrupted” by the older T cells in their midst. Restifo likens this phenomenon to quorum sensing in bacteria.
Another major hurdle of contemporary immunotherapy concerns finding appropriate targets for CAR-T cell therapy. CAR-T cell therapy is a specific form of adoptive T cell transfer in which T cells are removed from the patient, genetically engineered in the lab to recognize a cancer antigen, expanded to billions of copies, and then returned to the patient. To date, most successes with CAR-Ts have been obtained by targeting one antigen, called CD19, found on most B cells, including leukemia of B cell origin. This is the approach taken by researchers Carl June, M.D., at the University of Pennsylvania, and Michel Sadelain, M.D., Ph.D., at Memorial Sloan-Kettering Cancer Center, whose work has resulted in what look to be durable cures in a number of patients.
CD19 is in some ways a perfect target for CAR-T therapy because all CD19-positive cells in the body—both normal and cancerous—can be safely knocked out without causing a life-threatening condition. (The particular subset of B cells that confer long-lasting memory to infections, the plasma cells, are not CD19-positive.) But other antigens, unless they are truly tumor-specific, may not be suitable targets for this form of therapy.
Then there is the fact that even when a target seems to be appropriate in the lab, it might not translate to the clinic. Adrienne Long, an M.D./Ph.D. student at Northwestern University presented a fascinating talk aimed at understanding why CAR-Ts specific for GD2, an antigen found on sarcoma cells, work in the lab but have no activity in the clinic. Her research underscores the many subtle complexities that stand in the way of bringing CAR-T cell therapy to more cancer patients.
Viruses and Vaccines
On the vaccine front, this year’s focus was on oncolytic ("cancer-killing") virus-based vaccines. Researchers have long known that cancer remissions sometimes coincide with viral infections. One famous example involves a 4-year-old patient whose leukemia temporarily abated after he came down with chicken pox.
Several groups are now trying to take advantage of the tumor-lysing capacity of viruses to make better vaccines. Steven Russell, M.D., Ph.D., of the Mayo Clinic, and John Bell, Ph.D., of the University of Ottawa in Canada, each presented talks on how oncolytic viruses could be combined with vaccine approaches to produce more powerful results than vaccines alone. Russell talked about Amgen’s T-VEC, a GM-CSF transgene-containing version of an attenuated herpes virus, while Bell discussed research using a Maraba virus genetically modified to kill cancer cells but leave normal cells unharmed. Phase I/II trials of such oncolytic vaccines are being planned for later this year. This research is heartening given the recent and highly publicized failure of GlaxoSmithKline’s MAGE-3 vaccine.
Other promising research on cancer vaccines comes from researchers studying their use in the prevention context. Robert Schoen, M.D., M.P.H., of the University of Pittsburgh, presented research showing that the tumor-associated antigen MUC1 is a promising target for such a preventative vaccine. MUC1 is protein found on epithelial cells, such as those that line the intestine, that is overexpressed in many cancers, including colorectal and breast cancers. In mice models of colorectal cancer, vaccination against MUC1 is effective at preventing cancer. A pilot study of the vaccine in humans is scheduled to open soon at seven different sites. According to Schoen, the study will be the first immunoprevention trial based on a tumor antigen.
Elizabeth Jaffee, M.D., of Johns Hopkins University, presented evidence that vaccines may prove most effective when given along with immune-modulating agents like anti-PD-1. In the same session, Elizabeth Mittendorf, M.D., Ph.D., of MD Anderson Cancer Center, presented research suggesting that vaccines may be most effective in the adjuvant setting where there is minimal disease (such as after surgery). Mittendorf is the principal investigator of a phase III study of the HER2/neu vaccine called NeuVax in breast cancer, which is currently accruing up to 700 patients.
Given the many open questions that remain about how to maximize the promise of immunotherapy, it is important for people not to lose sight of the importance of continued support of research in this field. We are really only at the beginning of the immunotherapy revolution. More research, more funding, more commitment are all badly needed.