Study suggests why an immunotherapy drug works in some melanoma patients but not in others

In collaboration with the Press Association
Skin cancer cells under a microscope. Image: LRI EM unit.

Matching the size of a tumour to the body’s immune response could help doctors tailor immunotherapy treatments for melanoma patients whose disease has spread, according to a small US study.

Researchers found that patients who didn’t respond to treatment had an imbalance between the size of their tumour and how exhausted their immune cells were.  

The study, published in the journal Natureprovides clues that could help doctors identify patients whose tumours aren’t responding to certain immunotherapy drugs sooner, and possibly explain why.

Some immunotherapy drugs, such as pembrolizumab (Keytruda), work by blocking molecules on the surface of immune cells or tumour cells that act as a set of ‘brakes’ on the immune system. Blocking these molecules, including PD-1 which pembrolizumab targets, releases these ‘brakes’, turning immune T cells on to cancer cells.  

"The bigger the tumour, the more T cell 'reinvigoration' was needed by the drug.”  - Professor E. John Wherry, study lead

"We found that the size of the pre-treatment tumour determined how strong of a T cell response was needed in response to the drug to shrink a patient's tumour," said Professor E. John Wherry from the University of Pennsylvania who led the study.

He explained that, "the bigger the tumour, the more T cell 'reinvigoration' was needed by the drug.”

In a person with cancer, the T cells that may be able to attack cancer cells can become exhausted.

Over half of patients treated with immunotherapy for melanoma that has spread don’t see their tumours shrink long term.

The US team tested the blood of 47 patients with late stage melanoma, before and after treatment with pembrolizumab. They identified changes in circulating T cells and measured if exhausted T cells had been ‘reinvigorated’ by the drug.

The team then compared this with the size of the tumour at the start of treatment.

They found that patients whose tumour didn’t shrink had an imbalance between the intensity of T cell reinvigoration and the size of their tumours at the start of treatment.

"This research helps to answer some of the critical questions around why some patients don't respond to anti-PD1 inhibitors and why some do," says co-author Jedd Wolchok from the Memorial Sloan Kettering Cancer Center in New York.

Wolchok believes the study is important because it identified three different ways in which PD-1 blocking drugs can fail:

  1. If the drug doesn’t re-invigorate exhausted T-cells.  
  2. If an immune response is simply not strong enough for the size of the tumour.
  3. If the drug is off target.

Tim Elliott, Professor of experimental oncology at Cancer Research UK’s Southampton centre, called the study “interesting” and said it “suggests some possible ways of improving immunotherapy treatments”.

“The kind of scoring system that this study has developed could be used early on in immunotherapy treatment to predict how likely it is to work,” he added.

When the US team looked back through the data they could predict treatment failure 3 to 6 weeks in, 6 weeks earlier than in previous clinical trials.

“This is important as it could pinpoint quickly those in which the treatment isn’t working, so doctors can decide if another drug needs to be used,” said Elliott.

But Dr Edd James, associate Professor of cancer immunology at Cancer Research UK’s Southampton centre, pointed out that the “predictive power” generated by this study only came after treatment.

“Obviously, it would be better to be able to predict a patient’s response to a treatment before it begins, but the information gained from this study may be useful in identifying such a marker in the future.”

“How these findings can be applied to other cancer types now needs to be investigated,” he added.

References

Huang, A. C., et al. (2017). T-cell invigoration to tumour burden ratio associated with anti-PD-1 response. Nature DOI:10.1038/nature22079