By Anna C. Greene, PhD, NETRF Chief Scientific Officer
One of the most exciting developments in neuroendocrine cancer treatment in recent years has been the introduction of tarlatamab, a new immunotherapy. This drug is designed to help a patient’s immune system find and destroy cancer cells that express the DLL3 marker, a protein commonly found on aggressive neuroendocrine cancers such as small cell lung cancer (SCLC).
DLL3 is also found on extrapulmonary neuroendocrine carcinomas and on some lung neuroendocrine tumors (NETs). It is rare in low-grade (G1–G2) gastroenteropancreatic NETs, but more common in higher-grade (G3) NETs, especially pancreatic NETs.
But there has been a major challenge. Some people respond very well to tarlatamab, while others derive little or no benefit, and doctors have not had a reliable way to predict who will respond before starting treatment.
A new study from Shyamala Maheswaran, PhD, Justin F. Gainor, MD, and Daniel A. Haber, MD, PhD, and their research teams at Massachusetts General Hospital and Harvard Medical School, recently published in Cancer Discovery, offers a promising solution that uses a simple blood draw.
Why DLL3 matters
DLL3 is a protein found on the surface of many neuroendocrine cancer cells. Tarlatamab works by acting as a molecular bridge, binding DLL3 on tumor cells and CD3 on T cells, bringing the immune system into direct contact with the cancer.
In theory, if a tumor has DLL3, tarlatamab should work. In practice, it has been much more complicated.
Standard tissue biopsies often show that most SCLC tumors are DLL3 positive, yet more than half of patients will progress within six months of starting tarlatamab. This has made it very hard to know who should receive it, especially since the drug can cause serious side effects like cytokine release syndrome, which requires hospitalization.
A better window into the cancer, through blood
Instead of relying on tumor tissue collected months or years earlier, the research team turned to circulating tumor cells, also known as CTCs. These are whole cancer cells that break off from tumors and circulate in the bloodstream. They carry the same surface markers as the tumors that shed them.
Using highly sensitive microfluidic technology, the researchers were able to isolate these rare cells from blood samples and examine them one by one.
This allowed them to directly measure how many circulating cancer cells were actually carrying DLL3 at the moment a patient was about to start tarlatamab.
Two very different patient groups emerged
In 20 patients with advanced small cell lung cancer who were treated with tarlatamab, a striking pattern appeared.
Patients fell into two clear groups:
- DLL3-positive group
More than 25 percent of their circulating tumor cells carried DLL3. - DLL3-low group
Fewer than 25 percent of their circulating tumor cells carried DLL3.
The difference between these groups was dramatic.
What happened when patients received tarlatamab?
Among the 11 patients in the DLL3-positive group:
- 100 percent had clinical benefit
- More than half had tumors that actually shrank
- The rest had their cancer stop growing
Among the 9 patients in the DLL3-low group:
- Nearly 80 percent had their cancer continue to grow
- Only one had meaningful tumor shrinkage
In other words, simply counting the number of DLL3-positive cancer cells floating in the blood before treatment was enough to predict who was likely to benefit from tarlatamab.
Why tumor biopsies were misleading
The study also explains why standard tumor testing has not been helpful. When researchers looked at tumors cell by cell, they found that even within a single tumor, some cells had DLL3 and others did not. A biopsy might catch one area and miss another.
CTCs, on the other hand, appear to represent cancer cells actively spreading and driving disease, making them a more relevant target for treatment decisions.
What happens when the drug stops working?
Even among people who responded well, some eventually developed resistance. The researchers found two main reasons:
- Some tumors stopped making DLL3
The cancer essentially hid from the drug. - Some tumors still had DLL3, but the immune system became less effective
The patient’s T cells shifted into a less aggressive, more exhausted state and could no longer kill the cancer efficiently.
Importantly, even when DLL3 was lost, many cancer cells continued to express other neuroendocrine markers, such as SEZ6 and B7-H3, which are already being studied as new therapeutic targets. This means that patients who stop responding to DLL3-based therapy may still have other options ahead.
A glimpse into real-time tumor destruction
Another fascinating discovery came from early blood samples after starting tarlatamab. In some patients, researchers saw massive numbers of tumor fragments in the blood, pieces of cancer cells that had just been destroyed by an immune attack.
These fragments appeared only in patients who both:
- Had strong immune side effects, and
- Went on to have tumor shrinkage
This suggests that a simple blood test might one day help doctors detect, within days, whether a patient’s cancer is being effectively attacked.
What does this mean for the neuroendocrine cancer community?
Although this study focused on small cell lung cancer, DLL3 is also found in other aggressive neuroendocrine cancers.
Dr. Haber says, “We don’t yet know how well other neuroendocrine cancers will respond to tarlatamab or how predictive our blood test will be beyond small cell lung cancer, but we are very keen to find out.”
For patients and families, this research points to a future where:
- A simple blood test could help decide whether tarlatamab is worth trying
- Unnecessary side effects could be avoided in people unlikely to benefit
- Doctors could track whether a treatment is working or failing in real time
- New targeted therapies could be selected as tumors change
At NETRF, we view this work as a powerful example of how precision medicine, one of our Research Roadmap pillars, is becoming a reality for neuroendocrine cancers. Instead of guessing which therapy might help, doctors may soon be able to match each patient to the right treatment at the right time, using information from a single tube of blood.
This is exactly the kind of progress our community has been waiting for, which is why continued research and patient participation in clinical studies are so vital.