Stanford University scientists have found that erythropoietin (EPO), a protein identified nearly 40 years ago for its ability to stimulate the production of red blood cells also plays a surprising, critical role in dampening the immune system’s response to cancer. Their preclinical studies found that blocking activity of EPO converts formerly “cold” – or immune-resistant – liver tumors in mice into “hot” tumors teeming with cancer-fighting immune cells.
Combining EPO inhibition with an immunotherapy that further activates these immune cells against the cancer led to complete regression of existing liver tumors in the majority of treated mice, which lived for the duration of the experiment. In contrast, control animals survived only a few weeks. Although the work was carried out in preclinical models, there are strong indications that the protein, erythropoietin or EPO, plays a similar role in many types of human cancers.
“This is a fundamental breakthrough in our understanding of how the immune system is turned off and on in cancer,” said Edgar Engleman, MD, PhD, a Stanford University professor of pathology and of medicine. “I could not be more excited about this discovery, and I hope treatments that target the mechanism we uncovered will quickly move forward to human trials.”
Senior author Engleman, together with first author research scientist David Kung-Chun Chiu, PhD, and colleagues, reported on their findings in Science, in a paper titled “Tumor-derived erythropoietin acts as an immunosuppressive switch in cancer immunity.”
Successful cancer immunotherapy requires a patient to mount an effective immune response against tumors, but many cancers evade the body’s immune system, the authors explained.
For their reported study the team set out to investigate factors that govern antitumor immunity and tumor immunotype. Interestingly, the team noted, “EPO, a glycoprotein hormone known for stimulating red blood cell production, has recently been implicated in other biological processes involving resolution of inflammation.” Engleman added, “Research from more than a decade ago has shown that giving EPO to cancer patients with anemia to stimulate red blood cell formation accelerates the growth of the tumor.”
The connection was so striking that in 2007 FDA required a black box warning label on the drug cautioning against its use in people with cancers. Researchers also saw a clear correlation between patient prognosis and the levels of naturally occurring EPO and its receptor (EPOR) in the tumor. “Clinically, high expression of EPO is linked to poor prognosis in several cancers, including HCC, and accumulation of immunosuppressive cells such as regulatory T cells (Treg cells) and regulatory macrophages,” the team stated in their paper.
“Those old reports showed clearly that the more EPO or EPOR there was in tumors, the worse off the patients were,” Engleman said. “But the connection between EPO and cancer immunity was never made until now. In fact, it took a long time and a lot of experiments to convince us that EPO plays a fundamental role in blocking the immune response to cancer, because EPO is so well established as a red blood cell growth factor.”
The researchers were interested in the effect on cancer growth of a common immunotherapy targeting a molecule called PD-1 on immune cells called T cells. Binding to PD-1 blocks the ability of cancer cells to dampen the activity of T cells. Anti-PD-1 therapies are routinely used to treat many types of human cancers including melanoma, Hodgkin’s lymphoma and some types of lung cancer. In some cases they have transformed patient outcomes. But a large majority of tumors, including most liver, pancreas, colon, breast and prostate cancers are resistant to the treatment.
“Cancer patients with inflamed (T cell–rich) tumors, indicative of an active antitumor immune response, often benefit from immune checkpoint blockade (ICB) therapy,” the team noted in their research article summary. “However, despite the presence of tumor mutations that should theoretically trigger an immune response, most patients have noninflamed (T cell–deprived) tumors and do not benefit from ICB … the mechanisms that determine the immune cell profile or immunotype of tumors remain poorly understood.”
The researchers further pointed out that such noninflamed tumors are often “ … replete with immunosuppressive macrophages and neutrophils that hinder T cell priming, activation, and homing—critical processes for fostering antitumor immunity.”
Using somatic genome editing techniques the investigators created and compared several mouse models of liver cancer to help study how liver tumors develop and respond to treatment. “To investigate the basis for treatment failure, we examined spontaneous mouse models of hepatocellular carcinoma (HCC) with either an inflamed T cell–rich or a noninflamed T cell–deprived tumor microenvironment (TME),” the researchers explained in their paper. Each model recapitulates specific mutations, histology and the response to approved therapies found in subtypes of human liver cancers. Tumor formation was either induced by injecting a combination of DNA encoding proteins associated with liver cancer into the animals’ tail vein or by implanting liver cancer cells into the animals’ livers.
The researchers found that, similar to what has been observed in human liver cancers, some combinations of mutations led to the development of liver tumors that were largely ignored by the immune system, rendering them immune privileged, or cold. These tumors did not shrink when the animals were treated with anti-PD-1 because few T cells were present in the tumor.
In contrast to the cold tumors, other mutations led to hot or “inflamed” tumors that were replete with T cells. These tumors were highly sensitive to anti-PD1 treatment, which triggered the T cells to attack the cancer.
Unexpectedly, the cold tumors displayed elevated levels of EPO when compared with hot tumors. This increase is likely caused by the oxygen-poor microenvironment — a condition called hypoxia — prevalent in cold tumors. Hypoxia induces the production of proteins in cancer cells that, in turn, ramp up the production of EPO to create more red blood cells to combat low oxygen levels. “Hypoxia in tumors has been studied for decades,” Engleman said. “It just didn’t dawn on anyone, including me, that EPO could be doing anything in this context other than serving as a red blood cell growth factor.”
In their research article summary, the authors noted, “Using spontaneous preclinical models of HCC with distinct immunotypes, we show that the EPO/EPOR axis functions as an immunosuppressive switch in macrophages that maintains a T cell–deprived TME, thus posing a major barrier to effective antitumor immunity.”
The researchers also turned to existing databases to confirm that elevated levels of EPO are correlated with poorer survival of people with cancers of the liver, kidney, breast, colon and skin. They then tinkered with the ability of the tumor cells to make EPO and were surprised at what happened in the animals’ liver tumors.
They found that mutations that had led to the development of cold tumors instead caused hot tumors when the tumors were modified to be unable to make EPO. Conversely, hot tumors that had previously been successfully eradicated by the immune system thrived when they were engineered to make elevated levels of EPO.
Further exhaustive research showed that, in cold tumors, the tumor cells make and secrete EPO, which binds to receptors on the surface of immune cells called macrophages. “Tumor-derived EPO autonomously generates a noninflamed TME by interacting with its cognate receptor EPOR on tumor associated macrophages (TAMs),” the authors stated. The macrophages then switch to an immunosuppressive role, shooing away cancer-killing T cells and tamping down their activity. “Collectively, our data strongly support the notion that noninflamed HCC relies on EPO production to evade immune surveillance, with the macrophage EPO/EPOR axis being the primary immunosuppressive mechanism,” they added.
The importance of this EPO-moderated crosstalk between tumor cells and macrophages showed clearly when the researchers studied the combinatorial effect of simultaneously blocking the EPO signaling pathway and anti-PD-1 pathway.
In those experiments, no mice with cold liver tumors that were treated with control or with anti-PD-1 lived more than eight weeks after tumor induction. In contrast, 40% of mice with macrophages unable to make the EPO receptor lived for 18 weeks after tumor induction, the point at which the experiment was terminated. When anti-PD-1 treatment was given to mice lacking the EPO receptor, all animals lived for the duration of the experiment. “Removing either tumor-derived EPO or EPOR on TAMs leads to an inflamed TME and tumor regression independent of genotype,” the authors further pointed out. “Inactivation of EPO/EPOR reprograms macrophages to initiate a robust antitumor immune response, converting a noninflamed TME into an inflamed one.”
Added Engleman, “It’s simple. If you remove this EPO signaling, either by lowering the hormone levels or by blocking the receptors on the macrophages, you don’t just get a reduction in tumor growth, you get tumor regression along with sensitivity to anti-PD-1treatment.”
Engleman and his colleagues are now designing treatments targeting EPO signaling in human cancers. Non-specifically targeting the EPO protein could cause anemia, which Engleman speculates might be an acceptable trade-off for an effective cancer therapy. An alternative approach is to selectively block the EPO receptors on the surfaces of macrophages in the cancer. “I continue to be amazed by this finding,” Engleman said. “Not every tumor is going to respond in the same way, but I’m very optimistic that this discovery will lead to powerful new cancer therapies.”
The authors acknowledged that their study relied on murine models of HCC, which limits generalization, although they noted that “… the association between high EPO expression and poor prognosis across various solid malignancies suggests that similar mechanisms could contribute to the noninflamed immunotype and resistance to anti–PD-1 immunotherapy in other tumor types. On this basis, targeting the EPO/EPOR axis may have application for the treatment of solid tumors beyond HCC.”
