It’s well known that cancer cells can become drug resistant. When this happens, a previously effective treatment will stop working, causing major problems for patient care.

What is less well known is that cancer cells can become drug addicted. This puzzling phenomenon, which has been demonstrated in lab and animal models, and has been reported in patient case studies, suggests that certain tumours can sustain themselves through the very drugs designed to kill them.

If that occurs, it should in theory open a new avenue for treatment; experiments have shown that when you suddenly withdraw the drug, the addicted cells die en masse. However, this observation has yet to lead to clinical applications.

This may be set to change, however, following the publication of a new paper in Nature. The study, published in October, describes the mechanism behind cancer drug dependency, and points up a way that this vulnerability might be exploited in the clinic.

“When cells become resistant to particular drugs, they have to somehow compensate for the inhibitory effect of the drugs so they rewire their signalling pathways,” says Daniel Peeper, lead study author and professor at The Netherlands Cancer Institute . “If you acutely withdraw the drugs from these cancer cells, you disrupt the delicate balance between its rewiring and the potent action of the inhibitory drug. The cells will be unable to cope with the stress that emerges and they die – hence the word addicted.”

How cancer cells proliferate

As background, the propagation of cancer cells comes about as the result of mutations.

“Particular genes have acquired mutations, resulting in the production of mutant proteins,” says Peeper. “These act as engines of the cancer cells, allowing them to proliferate infinitely and metastasise in different environments. With recent sequencing techniques, it is quite straightforward to identify which mutations are present in cancers, and with developments in the pharma industry we’ve been able to use that information to develop precision drugs.”

While several dozen of these precision drugs have been approved and are used in the clinic, often prolonging the patient’s lifespan, such therapies do not commonly result in a cure.

“That is because cancer cells are highly flexible – they commonly suffer from dozens if not hundreds or thousands of mutations, which gives them great flexibility,” says Peeper. “Sooner or later they will switch their signalling circuits in such a way that they become insensitive to the inhibitory drug. That’s what’s commonly seen in patients who are treated with chemotherapy, but also with precision drugs.”

Disrupting the drug addiction

In the case of melanoma, Peeper’s particular area of study, the disease is commonly driven by a mutation in the BRAF gene. In these cases, it can be treated with a BRAF inhibitor (such as Zelboraf or Tafinlar) that attacks the mutant BRAF protein directly. While these drugs typically work for a while, eventually most tumours acquire resistance.

In order to understand the addiction mechanism at work, Peeper’s team decided to look at cancer cells that were both resistant to, and addicted to, this treatment. Using CRISPR-Cas9 technology, they painstakingly searched for the genes involved in drug addiction.

“In drug addicted melanoma cells, we observed that if we withdraw the drug most of the cells die,” says Peeper. “We designed this experiment such that each melanoma cell had a mutation in one particular gene, allowing us to select for cells that had survived the drug holiday. Most cells will die because they’re drug addicted, except for the ones that have an inactivating mutation in a gene essential for the drug addiction trait.”

The researchers identified three genes – ERK2, JUNB and FRA1 – which together formed part of a well-known signalling pathway.

“We showed that inactivation of any of these components will disrupt the drug addiction, and allow cells to completely bypass the drug holiday-induced death,” says Peeper. “This allowed us to uncover the pathway that’s essential for drug addiction.”

Intriguingly, the same pathway seemed to be operational in lung cancer cells with an EGFR mutation that had acquired resistance to an EGFR inhibitor, despite the fact these cells are addicted to a completely different treatment.

Alternating treatments

While these results are interesting in their own right, the really important question is how they might translate to the clinic. In one sense this is difficult to answer – since there have been no systematic studies in patients, we do not know what proportion of tumours are drug addicted, or are liable to regress on discontinuation of treatment.

“Previously, people have suggested that drug resistant tumours may be re-sensitised by a drug holiday,” says Peeper. “I think that’s a concept that works well in the laboratory, but there’s no clinical evidence that discontinuing the treatment will cause a massive and durable regression of tumours for most patients. Because of the common heterogeneity of tumours, there will be plenty of cells that rapidly overcome this drug addiction, and they will grow out again.”

His results suggest that, rather than taking a complete break from treatment, patients would be served better by receiving a second kind of therapeutic upon discontinuation of the first.

“We call this alternating treatment – the idea being to do the drug withdrawal when the tumour becomes addicted and to follow that immediately with a second treatment to increase the therapeutic benefit,” he explains.

In Peeper’s experiments, the secondary treatment chosen was a chemotherapeutic called dacarbazine – a melanoma drug that in itself shows little clinical benefit. However, despite being ineffective when taken alone, it strongly enhances the impact of the drug holiday.

Broader applications

Peeper’s team are currently looking into drugs other than dacarbazine, to work out which alternating treatments might be best placed to eradicate tumour cells. Beyond that, the question is whether a similar mechanism might apply to other kinds of tumour types.

“Are our findings much more broadly applicable to other tumour types treated with other inhibitors?” he says. “I don’t know the answer, but I could imagine that it is not limited to melanoma and lung cancer, because in essence many cancers one way or another rely on the pathway we’ve been studying here. Few tumour cells would be able to proliferate and survive if this pathway was inactive.”

It is entirely possible, then, that the researchers may have unearthed a finding that is applicable across many unrelated cancers. Further studies need to be conducted to determine whether this is the case – and beyond that, to determine the optimal treatments.

Although the clinical translation outlook of the paper is just a proof of concept, and should be greeted accordingly, the findings are undoubtedly exciting. Peeper says the paper has attracted a great deal of attention since publication, both from the media and from fellow scientists.

“I think the data are of potential clinical interest because drug resistance remains a formidable problem in the cancer clinic,” he says. “I hope this will pave the way for many other groups to look for the most optimal secondary treatment.”