Edwin Cuppen discusses tumour-agnostic drugs for cancer patients
Cancer medicine has traditionally focused on the site of a tumour, but in recent decades research has held out the promise of spotlighting its genetic background instead. The ultimate goal is to achieve more personalised treatment. Now that the first ‘tumour-agnostic drug’ has been authorised, that moment has finally arrived.
Some might describe it as ‘minimalist’, but in fact the office of Oncode Investigator Edwin Cuppen at the Hartwig Medical Foundation is just bare. Most professors’ rooms have shelves groaning under the weight of books and desks stacked high with papers, but here there are just three empty desks, a few chairs, and not a single cupboard. Prof. Cuppen ruefully explains that he has just moved in to this office. In fact, he divides his time between the Foundation in Amsterdam and his duties as Professor of Human Genetics at UMC Utrecht. He hasn’t got around to furnishing the place yet. For over 20 years, Edwin Cuppen has been using genetic technology to find chinks in tumours’ armour. He has witnessed the rise of personalised oncology at first hand.
I’m tempted to say “Wow – personalised cancer drugs, at last!”
“It has indeed been a long, hard slog – we’ve been at it for decades. Fundamental research is shedding more and more light on the genetic background and mechanisms of tumour formation. Years of research led us to the back doors that tumour cells use to evade the body’s control mechanisms and start dividing chaotically. These discoveries provided targets for potential drugs, but it takes quite a while for these to be developed. It takes even longer to find out whether they are safe and sufficiently effective to put on the market. The good news is that pharmaceutical companies currently have hundreds of drugs in the pipeline, each targeting a specific genetic defect. Many of these are independent of the type of cancer involved. There is a desperate need for this because, in many cases, cancer treatments still fail due to a lack of the right drugs.”
But one such drug is now being scrutinised by the EMA, the European body that grants marketing authorisation to new drugs. What kind of drug are we talking about here?
“It’s called Larotrectinib or Vitrakvi. This drug – a tropomyosin kinase inhibitor – is specifically for patients in which a kinase gene has fused with another section of DNA that activates it and triggers cancer. This genetic error is found in a whole series of cancers, ranging from soft tissue cancer to brain cancer and kidney cancer. This drug has now received conditional authorisation for use in all cancers in which this particular gene fusion is involved. It is the first tumour-agnostic drug. Here, ‘agnostic’ means that the tumour site has no significance. Using one drug to treat several types of cancers is nothing new, but initial authorisation was always for a primary tumour at a fixed site.”
“Larotrectinib has excellent efficacy; I just wish that every drug was as good. No fewer than 75 percent of patients with the relevant biomarker showed a response, half of them for a lengthy period of time. It seems that tumour resistance is slow to develop. I should stress that these were all patients who had run out of options in terms of routine treatment.”
“In fact, this particular mutation occurs in about 0.5 percent of all tumours. That may not sound like a lot, but – given the large numbers of cancer patients – it really does make a difference for a significant number of patients. Clearly, with more drugs like this, we will be able to keep improving survival rates of many more cancer patients.”
What else is in the pipeline?
“Many pharmaceutical companies currently have PARP inhibitors in various stages of clinical trials. Some of these will, no doubt, appear on the market before too long. PARP is an enzyme that is involved in the repair of ‘nicks’. These are minor errors in the DNA which lead to more serious genetic problems during cell division. PARP inhibitors are mainly effective in cells in which another DNA repair mechanism is already defective and that, as a result, are dependent on PARP. The inhibitors, therefore, prevent these DNA errors from being corrected. This causes the damaging effects of these errors to accumulate, ultimately resulting in the death of the tumour cell. These drugs are highly effective, and are registered for use against ovarian cancers that involve BRCA-mediated homologous recombination deficiency. However, homologous recombination deficiency is also present in 10 to 15 percent of breast-, pancreatic-, prostate- and bladder cancers. Based on the molecular mechanism involved, it is reasonable to expect that these drugs might also be effective in patients with these types of tumours. This possibility is currently being investigated in a range of studies, which are also making use of the Hartwig Medical Foundation’s database.”
“Another mechanism found in many different cancers is microsatellite instability, which is caused by impaired DNA repair. This mechanism was already known to be involved in certain types of bowel cancer, including hereditary forms of this disease. However, this defect appears to occur much more widely, in about six percent of all patients on average. In addition, it appears to provide the basis for a good response to tumour-site independent immunotherapy.”
So, are we abandoning the idea that the tumour site dictates the type of treatment used?
“No, the exact site of a tumour continues to be an important factor. This is because the genetics are not the whole story – the microenvironment also plays an important part. Some tumour sites are difficult for drugs to access. For instance, they may be protected by stromal cells or they may not be accessible to the immune system. I can well imagine that future oncologists will consider both of these aspects when reaching a diagnosis – what is the genetic background, and what are the features of the microenvironment?”
“We may need to consider new combination therapies – one drug to improve accessibility, followed by one that actually attacks the tumour. We have now reached the point at which we can screen a tumour’s genetic background quickly and – almost – completely. Its microenvironment is a bit trickier, however, since we don’t always know what to look for. We need to amass much more fundamental knowledge to properly understand the interactions between tumours and their environment.”
Is a better treatment for metastases also within our grasp?
“Traditionally, the search for new cancer drugs focuses on the primary tumour. When a tumour is surgically removed, the surgeon takes a tissue sample. Researchers can then use this sample to investigate the cause of the cancer. Metastases often develop in many different sites in the liver or in the bones. This makes surgical removal much more difficult, which in turn means that far fewer samples are available for research purposes. In practice, while current treatments can cure many primary tumours, any associated metastases are much more difficult to treat.”
“However, with the help of 50 hospitals and studies – coordinated by the Center for Personalized Cancer Treatment – we were able to obtain thousands of samples of metastases from patients who were eager to do their bit for science. That’s really amazing, as these people did not stand to benefit directly, indeed they were exposing themselves to even greater risks.”
“We have sequenced the complete genome of more than 4,000 metastases documented in the Hartwig Medical Foundation database and published our results in Nature. Our goals included an attempt to identify any known, treatable mutations. It turned out that in no less than 62 percent of the patients we found changes in the DNA that could be treated using existing (about half of all patients) or experimental drugs. In some cases, this involved drugs that had already been approved for the type of cancer affecting the patient in question. In other cases, these were drugs that only had market authorisation for use against other types of tumours – known as ‘off-label’ applications.”
“There is, of course, no guarantee that these people will actually benefit from such drugs. All sorts of things can throw a spanner in the works. For instance, some tumours may be difficult to access or the molecular processes involved may vary from one cell type to another. Now we have to repay these people’s trust, and get to grips with this matter.”
It sounds as if we need to rethink the classification of all kinds of existing cancer drugs?
“Yes, indeed. Many drugs may have much wider applications than those for which they are currently registered. In 2016, the Center for Personalized Cancer Treatment set up the Drug Rediscovery protocol in the Netherlands. This was partly prompted by the need to better structure off-label cancer drug use and partly because we were increasingly identifying potentially promising leads for patients whose tumours had undergone whole genome sequencing. In the past, patients have occasionally been given off-label treatments. However, these were usually restricted to single cases that often were not properly documented. In recent years, we have been using the new protocol in a methodical search for off-label applications. If we find that a given treatment is effective in a larger group of patients (or even if it is not, as negative results are also important), we publish our results and take appropriate action. We’ve already had our first success. Nivolumab, a drug used to treat skin and lung cancers, can now be used more widely to treat all patients with tumours that have microsatellite instability. At a certain point in time, that drug will be reimbursed by health insurers. On average, we find that 34% of patients do benefit from off-label treatments across a range of targeted and immuno therapies.
Has all that genetic sequencing work brought us any closer to finding out why some tumours spread and others don’t?
“Surprisingly, we did not find any genetic changes that could explain why tumours spread. Rather predictably, most of the metastases we studied had the same driver mutations as the tumour from which they came. The few cases in which new driver mutations were found mainly involved resistance to treatments that these patients had previously undergone.”
“Many scientists now think that metastases are actually not that special – genetically speaking. Solid tumours are constantly shedding cells into the bloodstream. Whether or not these cells ultimately produce new, metastatic tumours somewhere else is largely a matter of chance. For instance, the immune system would have to slip up and allow a circulating tumour cell to escape, after which that cell would have to settle in a spot where the conditions are exactly right for it to grow. Sadly, this means there is little chance of finding special drugs that can completely suppress the process of metastasis.”
What impact will all these changes have on clinics?
“Doctors currently work with standards of care, based on guidelines. That means starting with the drug that is effective for the largest group of patients with that specific tumour. If it doesn’t work, then they work their way down the list of treatment options until they get a hit. Some genetic testing does take place, but this tends to focus on only a few fragments of DNA that are known to be relevant to the tumour in that specific site.”
“I would like to propose that we get started right away on systematically sequencing the entire genome of the tumour. The list of genetic errors that we need to screen for is becoming longer and more complex by the day. You could, of course, deal with that by developing new and specific tests, as needed. However, the smart approach would be to simply read the entire DNA sequence straight away. At the Hartwig Medical Foundation, we can now do that very rapidly, and with diagnostic precision. As a result, all potential standard of care and experimental targeted treatment options are catalogued and can be used for treatment decision making. Furthermore, the complete sequence is also immediately available for research for improving cancer care. For instance, it can be used to find similarities between patients for whom a given treatment does not work. That would enable us to further refine biomarkers for the stratification of future patients, to reduce levels of over-treatment and to achieve savings with regard to expensive drugs. This would unify diagnosis and research while, in the long term, improving healthcare and keeping it affordable.”