Although it is one of the most aggressive solid tumors in humans, NUT carcinoma responds better to treatment in some patients than others. But because it is rare — with only 20 to 30 cases diagnosed annually in the United States — doctors have had little way to predict which patients are likely to benefit from standard treatment and which are not.

By analyzing data from more than 100 patients from 17 countries, a team led by Dana-Farber and Brigham and Women’s Hospital investigators has for the first time established a framework for assessing a patient’s likely response to therapy. Their model, described in a recent paper in the journal JNCI Cancer Spectrum, places patients in one of three prognostic categories based on the part of the body where the cancer arises and the specific molecular irregularity in their tumor cells. The model enables physicians to connect patients with therapies most likely to be helpful for them, be they standard therapies or agents undergoing testing in clinical trials.

The potential of novel agents for the disease is evident in a second paper by Dana-Farber researchers in JNCI Cancer Spectrum. In an early-phase clinical trial, a compound that targets a key vulnerability in NUT carcinoma cells proved to be safe and showed preliminary activity against tumor cells. The trial, which included the largest group of patients with NUT carcinoma to participate in a single study, makes a convincing case for further trials of this approach to treatment, the study authors say.

What is NUT carcinoma?

NUT carcinoma — formerly known as NUT-midline carcinoma — is a type of solid tumor that can form anywhere in the body but often occurs in midline structures such as the head, neck, or lungs. If the tumor is confined to a single space and hasn’t metastasized, it can potentially be cured by surgical removal.  Unfortunately, such cases are the minority. Some patients respond well to chemotherapy, but those responses rarely are lasting. The disease generally recurs, and the median survival time following diagnosis is less than seven months.

The disease is defined by a molecular mishap known as a translocation, in which the gene NUTM1 is inserted into the wrong section of the genome. The result is a fusion gene, an amalgam of NUTM1 and whatever gene happens to be adjacent. That, in turn, gives rise to a fusion protein made up of NUT (the protein encoded by NUTM1) and the protein encoded by NUTM1’s new neighbor in the genome.

In NUT carcinoma, NUTM1 is most often fused to the gene BRD4, creating the BRD4-NUTM1 cancer gene, but can also be bound to a variety of other genes, including BRD3, NSD3, ZNF532, and others.

Targeting a protein

In the new study, investigators asked whether the initial site of a NUT carcinoma tumor and the specific fusion protein within its cells was related to how well patients responded to therapy. To find out, they analyzed tumor tissue from 124 patients with the disease and correlated the data against their treatment histories.

“We identified three statistically distinct risk groups,” says the study’s senior author, Christopher French, MD, of Brigham and Women’s Hospital. “Patients whose primary tumor occurred outside the thorax and whose tumor cells had NUT-BRD3 or NUT-NSD3 fusion proteins had the best response to therapy and the longest survival. Patients whose primary tumor was in the thorax had the worst prognosis. And patients with non-thoracic tumors that had the BRD4-NUT were in the middle.

“Our findings, if validated, indicate that molecular testing to characterize NUTM1-fusion oncogenes may have significant value in guiding treatment decisions.”

The second study, published together with the prognostic model, reported encouraging results in a clinical trial of the agent molibresib in 65 patients with NUT carcinoma and other solid tumors.

Molibresib is a compound that targets a family of proteins with a section called a “bromodomain.” Such proteins “read” certain chemical attachments on DNA and thereby influence how active genes are. The bromodomain family includes BRD3 and BRD4, two of the proteins that NUT partners with to form the fusion proteins that spur NUT carcinoma growth. (The fusion proteins accomplish this by ratcheting up the activity of a gene called c-MYC, one of the prime engines of cancer cell growth.) Notably, Dana-Farber biologists were among the first to develop potent inhibitors of bromodomain proteins; subsequent refinement of these compounds led to drugs suitable for clinical testing.

In laboratory samples of NUT carcinoma cells, scientists at Dana-Farber and Brigham and Women’s had shown that blocking the ability of NUT-fusion proteins to attach to DNA with bromodomain inhibitors can bring cell growth to a halt. These findings led investigators to open trials such as the molibresib trial, designed to test an agent that directly targets the NUT-fusion cancer protein.

In the trial, 65 patients with solid tumors — including 19 patients with NUT carcinoma — received molibrisib. At doses where there was evidence that the agent was reaching its target proteins within tumor cells, it was found to be safe.

Of the 19 participants with NUT carcinoma, four had a confirmed or unconfirmed partial response (a decrease in the size of the tumor or the extent of the cancer), and eight had their disease remain stable. In some cases, responses or instances of stable disease well exceeded six months, considered an important advance in a relentlessly aggressive and recalcitrant disease.

“Our results provide evidence that molibresib engaged its target protein and had preliminary antitumor activity,” says the study’s senior author, Geoffrey Shapiro, MD, PhD, director of the Early Drug Development Center at Dana-Farber. “These results provide compelling evidence of the potential of drugs that inhibit bromodomain protein activity in patients with NUT carcinoma.”

This article was originally published on January 14, 2020, by Dana-Farber Cancer Institute. It is republished with permission.