Turning Off Metastasis in Breast Cancer

Share this post on:

Immunotherapy has revolutionized cancer treatment, but its effectiveness hinges on whether immune cells penetrate and persist within tumors. However, many cancers are immunologically cold, meaning that they do not trigger immune responses.1 Turning these cold tumors into hot ones is pivotal for improving immunotherapy effectiveness, and Lingyin Li, a chemical biologist at Stanford University, may have found a key on-off switch. 

Photo of Lingyin Li

Lingyin Li believes that the ENPP1 enzyme is a targetable weak point in cancer cells.

In a study recently published in the Proceedings of the National Academy of Sciences, Li and her team at Stanford University showed that an enzyme named ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1) drives breast cancer immune evasion and metastasis.2 They also reported that inactivating ENPP1 activity in mice almost eliminated metastasis. Their findings suggest that targeting ENPP1 may provide therapeutic benefits for late-stage patients.

Under normal circumstances, the human body has multiple mechanisms to prevent uncontrolled proliferation and tumor formation. The stimulator of interferon genes (STING) protein mediates one such mechanism. STING drives pro-inflammatory interferon cytokine production when double-stranded DNA is present in the cytosol, something that only happens in cancer cells. In 2013, a synthetic STING agonist, 5,6-dimethylxanthenone-4-acetic acid (DMXAA), showed tremendous anticancer efficacy in mice but failed in human clinical trials because the murine and human versions of the protein differed too much.3

DMXAA’s failure spurred Li, at the time a postdoctoral fellow in systems biologist Tim Mitchison’s group at Harvard Medical School, to investigate natural mechanisms for STING activation in humans. STING’s main endogenous agonist is 2′-3′-cyclic GMP-AMP (2’3′-cGAMP), which is regulated, like other cyclic nucleotide second messengers, by phosphodiesterases. However, no one knew which ones. A chemist by training, Li saw this as an opportunity to purify 2’3′-cGAMP’s phosphodiesterase. “Traditionally, what chemists do really well is purify enzymes from bulk sources. So, I rolled up my sleeves, went into the cold room, and did not come out for months until I had purified ENPP1.”4 

Li hypothesized that blocking ENPP1 activation would allow 2’3′-cGAMP to continue to activate STING, leading to a more robust and persistent immune response and promoting hot tumors. However, if tumor cells showed upregulated ENPP1 activity, the opposite would occur, resulting in cold tumors. By analyzing patient data compiled by the Molecular Taxonomy of Breast Cancer International Consortium, Li and her team confirmed that breast cancer patients with higher levels of ENPP1 mRNA showed significantly worse disease-free survival rates. Further, they found that patients with stage IV metastatic disease had significantly higher ENPP1 RNA expression than patients with stage III disease, leading them to further investigate ENPP1’s role in tumor growth and metastasis.2

Cancer cells cannot so easily mutate out of this weakness, making [ENPP1] their Achilles’ heel.
—Lingyin Li, Stanford University

To that end, Li and her team injected mice with breast cancer cells engineered to overexpress ENPP1. However, there was a twist. Researchers injected some of the mice with cells that produced fully functional ENPP1, while others received cells with inactivated ENPP1. This latter group showed lower rates of tumor growth and virtually no lung metastasis. Metastasis returned when both ENPP1 and STING were inactivated, showing that ENPP1’s antimetastatic effects depended on the STING pathway.2 

ENPP1 serving as an on-off switch for metastasis is striking given the complexity of cancer. “It would be silly to think that just one gene dictates everything that happens in tumorigenesis,” Li said. Yet, she and her team did not notice any other signaling changes in response to ENPP1 inactivation. “Metastasis requires chromosomal instability, but in reaching this state, DNA inevitably leaks into the cytosol and activates the STING pathway,” Li explained. “As a result, cancer cells cannot so easily mutate out of this weakness, making [ENPP1] their Achilles’ heel.”

All of these findings make ENPP1 a particularly attractive target for enhancing STING signaling for anticancer therapeutics. “Most of the focus in the immunotherapy field has been to [activate] STING with direct agonists. But it is challenging to control the extent of this activation and avoid toxicity associated with systemic administration,” noted John Wilson, an immunoengineer at Vanderbilt University who was not involved with this study. “Through ENPP1, it is possible to leverage natural spatiotemporal control mechanisms for STING activation.” Wilson further opined that combination approaches using ENPP1 inhibitors in conjunction with STING agonists have a lot of promise for boosting overall therapeutic potency.

Finally, ENPP1 is unique among phosphodiesterases in that it is a cell surface transmembrane protein with a catalytic domain that faces the outside of the cell. This allows it to degrade extracellular 2’3′-cGAMP that has leaked out of a cancer cell before it alerts immune cells but also renders it vulnerable to small molecule targeting. Noting that enzymes are good targets because they naturally bind small molecules, Li said that ENPP1 is as easy to work with as the phosphodiesterase targeted by the erectile dysfunction medication Viagra. Li hopes that their work can help researchers develop something just as popular and widely used—the anticancer similitude of Viagra.


  1. Galon J, Bruni D. Approaches to treat immune hot, altered and cold tumours with combination immunotherapies. Nat Rev Drug Discov. 2019;18:197-218.
  2. Wang S, et al. ENPP1 is an innate immune checkpoint of the anticancer cGAMP–STING pathway in breast cancer. Proc Nat Acad Sci. 2023; 120(52):e2313693120.
  3. Kim S, et al. Anticancer flavonoids are mouse-selective STING agonists. ACS Chem Biol. 2013;8(7):1396-401. 
  4. Li L, et al. Hydrolysis of 2’3′-cGAMP by ENPP1 and design of nonhydrolyzable analogs. Nat Chem Biol. 2014;10(12):1043-48.
Share this post on:

Leave a Reply

Your email address will not be published. Required fields are marked *