A man and woman indoors smiling
A Dynamic Duo: Dorothy Sipkins, MD, PhD, and P. Kelly Marcom, MD, are co-PIs on a clinical trial of a new drug for breast cancer patients whose disease has spread to the bone.

New Metastatic Breast Cancer Drug Trial Launches


Duke Cancer Institute investigators Dorothy Sipkins, MD, PhD, associate professor of medicine, pharmacology and cancer biology, and P. Kelly Marcom, MD, professor of medicine, have just launched a proof-of-concept clinical trial of a new drug for hormone receptor positive breast cancer patients whose disease has spread to the bone.

The investigational therapy, GMI-1359, targets both E-selectin and CXCR4, normal inflammatory molecules that also appear to play a role in tumor trafficking and metastasis to the bone.

The drug, discovered and invented by glycobiologist John Magnani, PhD, Senior Vice President and Chief Scientific Officer at GlycoMimetics, Inc., intersects with years of preclinical research by Sipkins, first in leukemia and then in breast cancer.

The new therapy, which co-principal investigators Sipkins and Marcom are calling “a micro-environment targeted therapy,” has the potential to improve treatment of patients at risk of cancer metastasis to the bone, or whose cancer has already “micro-metastasized” to protective pro-dormancy niches in the sinusoid capillaries (sinusoids) of the marrow.

“The Sipkins therapy, for short,” said Marcom, crediting Sipkins’ “world class bench science and translational work.”

“John Magnani has been researching E-selectin for years,” said Sipkins. “My work showing the role of both E-selectin and the SDF-1 receptor, CXCR4, in leukemia cell trafficking to the bone marrow [published in Nature, 2005] intrigued him, and he considered the possibility that a drug that inhibited both molecules might be very useful.”

In that study, Sipkins showed that CXCR4 (with its chemokine SDF1) and E-selectin played a role in trafficking and adhesion of leukemic cells into the sinusoids.

After her discoveries in leukemia, she pivoted to breast cancer.

“I became obsessed with these two molecules, E-selectin and CXCR4,” said Sipkins. “The first question we had was if breast cancer cells were using the same gateways to get into the bone marrow.”

When she learned that breast cancer cells, unlike leukemic cells, didn’t need CXCR4/SDF1 to enter the sinusoids, she become more focused in on E-selectin.

Sipkins then asked Magnani for access to a drug he was developing in the hematology space to inhibit E-selectin — GMI-1271/uproleselan.

Woman in lab coat in front of two computers
Dorothy Sipkins, MD, PhD, uses laser scanning confocal microscopy to “spy”on breast cancer in real time as it spreads into the bone. (photo by Ken Huth)

In mouse studies, a team in Sipkins lab, led by Sipkins and senior research associate Trevor Price, PhD, found that the drug successfully prevented breast cancer cells from entry into the bone marrow, validating the hypothesis that it was E-selectin that was causing breast cancer cells to slow their roll along the body’s vascular highway and enter the sinusoids.

Meanwhile, in the same study (published in Science Translational Medicine), they identified CXCR4/SDF1 as an important mechanism for anchoring the micro-metastatic breast cancer cells in the sinusoids.

Giving the mouse the CXCR4 inhibitor plerixafor (AMD3100) — an agent FDA-approved for use in human bone marrow donors to push stem cells into the bloodstream for harvesting — appeared to force dormant breast cancer cells from their niche in the sinusoids back onto the vascular highway where they might be more receptive to cancer therapies or targeting by the immune system itself.

Her team has since found that GMI-1359 also forces breast cancer cells from the niche, but for longer periods of time than AMD3100. Importantly, after treating mice for extended periods of time with GMI-1359, the team found no follow-on metastasis to other organ sites.

“One concept in tumor biology is that if epithelial cells lose their attachment to the tissue microenvironment for an extended period of time, they’ll start to undergo what we call programmed cell death, apoptosis. They’ll essentially activate a suicide program,” Sipkins explained. “So, we think that if we launch these tumor cells out of the bone marrow with the CXCR4 inhibitor and then we prevent them from re-entering the bone marrow with the E-selectin inhibitor, then the cells might just stay circulating in the bloodstream and potentially die.”

The dual-inhibitory drug, in this way, might even make breast cancer cells more amenable to immunotherapies (including checkpoint inhibitors) that have not traditionally been effective in hormone receptor positive (HR+) breast cancer, Marcom added.

The bone is the most common site for late relapse in HR+ breast cancer.

“Even at very early stages, these roaming cancer cells are making their way out of the breast and into the bone marrow,” Marcom explained, “and this can happen even before patients are diagnosed with breast cancer.”

In fact, clinically silent bone marrow micro-metastases can be detected in 30% of breast cancer patients with stage I to III disease. So, while a patient may go through therapy and appear “cured” — i.e. show no evidence of disease — there can be dormant breast cancer cells already settled in the sinusoids of the bone marrow.

Man in white doctor's coat smiling
Kelly Marcom, MD

Bench to Bedside

With HR+ breast cancer alone accounting for some 70% of all breast cancer diagnoses, Sipkins’ discovery of the mechanisms by which these cancer cells get to into these pro-dormancy niches and how they might potentially be excised out before they can grow and multiply, is significant.

“At baseline, the sinusoids always look like inflamed blood vessels. And that’s what makes them special; that’s what allows tumor cells to dock there,” explained Sipkins. “What we think is that the breast cancer cells come in through the sinusoids, and they can hang out there for a long time doing nothing. Then, they migrate to these areas that are more removed from the sinusoids and closer to bone cells — an environment where they’re more likely to proliferate. The goal is to cut them off at the pass earlier in the disease; when they’re still hanging out in the sinusoids and haven’t yet got themselves up to no good.”

It’s impossible to predict, however, just when they might “get up to no good.”

“We have no clear understanding of why these dormant cells wake up and start growing,” said Marcom, noting that these breast cancer cells could lie dormant for five, 10, 20 or more years, with the longest he’s seen, 35 years. “In the early 1990’s, the ‘angiogenic switch' was all the rage and that is likely part of the explanation. However, given what we know now, the mechanism is clearly complex.  Another likely factor is that these dormant cancer cells are kept at bay by the effectiveness of the immune system until something changes and then they quickly grow some, get more genomic instability, then they take off. This is arguably the most important l final problem to be solved in treating breast cancer.”

Marcom, a medical oncologist who directs the Breast Cancer disease group, is encouraged by the possibilities of this new therapy.

“I haven’t been this excited about a new potential therapy since the CDK cell cycle inhibitors came along” he said. “It’s really a novel mechanism that addresses, no pun intended, a novel niche of therapy.”

Six patients with known HR+ breast cancer metastasis to the bone — stage 4 — will be enrolled for the single-site Phase 1b clinical trial of GMI-1359, to be held at Duke. Qualifying participants must be stable (or with slight progression) on anti-hormonal therapy as well as a CDK4/6-inhibitor (abemaciclib, palbociclib, or ribociclib) — the current standard-of-care therapy with for metastatic hormone driven breast cancer. 

Participants will receive increasing doses of the study agent once a month over approximately three months, then three successive daily doses at the highest tolerated dose.  While the primary focus of the trial is safety of the drug, the effects on normal blood cells, immune system markers, and cancer cell mobilization will also be analyzed. 

“This trial is going to be a proof of concept to see whether or not we see a bump in tumor cells circulating in the vasculature after a dose of the agent,” explained Sipkins. “We don’t know for sure, however, if we’ll have to see an elevation in the circulating tumor cells for this drug to be active in breast cancer patients. It’s possible that we’ll find that, even without having to leave the bone marrow, the inhibitor will render the cancer cells more sensitive to other treatments or put them in a more pro-apoptotic state.”

An exploratory objective of the trial will be examining how normal white blood cells in both the bone marrow and in circulation may change with this therapy.

If the study proves that the agent is safe, tolerable and there’s some signal that at the maximum dose that there are biomarkers to suggest the drug is hitting its targets (E-selectin and CXCR4), then a phase 2 trial in breast cancer may be the next step.

“Ultimately, the goal is to get this into early stage patients; patients who have finished conventional adjuvant treatment and are on endocrine therapy but for whom there’s evidence of residual tumor in the body; for example, a supersensitive circulating tumor cell DNA assay,” said Sipkins. “They’d then get treated with GMI-1359 in order to purge this micro metastatic disease. We’d also like to be able to use this once patients get to the metastatic setting where we treat them and try to control it.”

Sipkins’ office whiteboard is covered with formulas, drawings and equations. She said she can’t believe she’s been at this for nearly 30 years.

“As a physician scientist, you spend your life obsessing on this stuff and you hope that it has some relevance,” she said. “I hope this comes to something good.”

This page was reviewed on 12/09/2020