In the late 1990s, oncologists found a winning combination for treating ovarian cancer. When given alongside a drug, called paclitaxel, that blocks cell division, the platinum-based chemotherapy cisplatin succeeded in shrinking and killing off tumours left after surgery in 70–80% of patients. Since then, this combination has been the treatment of choice for high-grade serous ovarian cancer, the most common and deadliest form of the disease. ‘The upfront response rates to these therapies are so high that if you had to pick a single drug, hands down, it would be platinum,’ says Anil Sood, a gynaecological oncologist and researcher at the University of Texas MD Anderson Cancer Center in Houston.But platinum has a resistance problem. About 10–15% of patients have tumours that do not respond to cisplatin. And even when tumours do initially respond, more than 80% of them return, and resistance will eventually emerge. ‘Most patients who have relapsed disease will develop platinum-resistant disease at some point,’ Sood says. Part of Nature Outlook: Ovarian cancer If the cancer recurs in the six months after the end of chemotherapy, it is considered resistant, and subsequent treatment will rely on non-platinum-based options. Patients might be offered an alternative chemotherapy alone or in combination with drugs that inhibit blood-vessel growth, or be enrolled in a trial of a new treatment. But these approaches shrink tumours only about 10–15% of the time; on average, people with platinum-resistant ovarian cancer survive for no longer than a year.Many alternative approaches for tackling platinum-resistant tumours have arrived in the past few years. Most are built on a growing understanding of what makes these malignancies so stubborn, such as their enhanced ability to undo DNA damage and promote blood-vessel growth. ‘We’re starting to really mine the knowledge that’s been developed over the last decade and a half,’ says Elise Kohn, an oncologist at the US National Cancer Institute (NCI) in Bethesda, Maryland.Advances in understanding the cancer’s biology have led to a slew of clinical trials — but the results have often been disappointing. ‘Most of the past trials in platinum resistance were negative,’ says Stéphanie Lheureux, a medical oncologist at the Princess Margaret Cancer Centre in Toronto, Canada. ‘It’s a very challenging disease to treat.’Unpredictable responsePlatinum-based drugs work by attaching to DNA and causing breaks in the molecule. These breaks interfere with the cell’s ability to copy its genome, killing the cell. But platinum-induced DNA damage also activates cells’ self-repair mechanisms.One of these processes, called homologous recombination, takes advantage of the fact that somatic cells carry two copies of each chromosome (with the exception of the Y chromosome) and replace broken parts of a gene on one chromosome with its intact counterpart on the other. The genes BRCA1 and BRCA2 play major parts in this process. Although a mutation in either gene contributes to the risk of developing ovarian cancer, people who carry mutations in either one lack a fully functioning DNA repair kit and can therefore be expected to respond well to platinum-based therapy.Oncologists can further assess a tumour’s ability to repair its DNA by quantifying defects in its genome, such as missing or rearranged chunks or the accumulation of certain types of mutations. The more of these genomic scars there are, the more likely it is that homologous recombination will not work well — which is good news for platinum-based therapies. For people who have low levels of genomic scarring and no BRCA mutations, the story is usually bleaker, says Katherine Fuh, an oncologist and researcher at Washington University in St. Louis, Missouri. We know that those people are a more difficult to treat group, and would usually develop platinum resistance, she says.However, genomic scarring and BRCA mutations are not precise predictors of a person’s response to platinum-based chemotherapy. For starters, Kohn cautions that although tests for these markers often correlate with responses to platinum-based therapy, that is not their intended use. Tests to detect genomic scarring — two of which have been approved by the US Food and Drug Administration — are designed to find people who are likely to respond to treatment with a class of drugs called poly(ADP-ribose) polymerase (PARP) inhibitors, which interfere with another method of DNA repair. Ovarian-cancer genomes can also change a lot over time, meaning that tumours could acquire mutations that alter their ability to perform DNA repair after a person has been tested. Epigenetic modifications that affect the cells’ capacity for DNA repair might also pass undetected. Katherine Fuh (right) is an oncologist at Washington University in St. Louis, Missouri.Credit: Matt Miller/Washington Univ. School Of Medicine
Given that current measures of DNA repair do not strictly align with an individual’s response to platinum or other chemotherapies, David Tan, an oncologist at the National University Cancer Institute in Singapore, thinks that researchers need to find other molecular features of platinum sensitivity that can be layered on top of existing tests.In March, Tan and an international group of researchers used a microscopy-based method to measure one potential marker of platinum responsiveness: a protein involved in homologous recombination, called RAD511. The team measured RAD51 levels in tissue samples from people who had received platinum-based chemotherapy alone as their first-line treatment. They found that individuals with the lowest RAD51 levels had better outcomes and experienced cancer recurrence later than those with the highest levels. Because the tumour samples had been exposed to nothing but platinum, Tan thinks that low RAD51 activity might be a marker indicating that platinum resistance is less likely to develop. He suggests that clinicians could use RAD51 when deciding whether to combine another drug with platinum-based chemotherapy during the first round of treatment, to start a second maintenance drug once chemotherapy is complete, or to encourage a patient to enrol in a clinical trial.The search for biomarkers is complicated by limited access to tumour tissue. Biopsies are expensive and are rarely taken after treatment has begun. When repeat biopsies are possible, drawing conclusions is complicated because changes observed in tissue from one tumour site might not be present elsewhere. Some researchers are hopeful that ovarian organoids — miniature versions of the structures grown in a dish — or circulating tumour cells taken from the bloodstream could provide ways to track tumour responses to drugs and changes to DNA repair genes in real time, but these tools are just starting to be established for ovarian cancer.Combinations for the win?Platinum resistance is so hard to predict and treat because it has no single cause, Sood says. In addition to exhibiting enhanced DNA repair, tumour cells might ramp up processes that improve cell survival, make higher levels of molecular pumps to shunt drugs out, or produce proteins that sequester platinum away from the genome. Any combination of these mechanisms can be at play in an individual.Decades of clinical trials have tested drugs that target these individually. But because ovarian tumours are constantly adapting, going after one resistance mechanism is not enough. ‘Unless you take a broad approach, you’re circumvented by other strategies,’ Sood says. The most promising approaches are therefore those that attack platinum-resistant tumours from several angles.For example, in a 2014 study, researchers combined paclitaxel with a monoclonal antibody called bevacizumab, which blocks blood-vessel formation by inhibiting vascular endothelial growth factor (VEGF). The combination increased the survival time of people with platinum-resistant tumours by three months compared with those on the chemotherapy alone2. The ways in which the antibody works have nothing to do with platinum resistance as such, Sood says, but the combination has nonetheless become a standard for people with resistant tumours.Laboratory studies3,4 showing that VEGF inhibitors reduce the production of BRCA and RAD51 proteins have also led some research groups to combine the compounds with PARP inhibitors — the hope being that the addition will make platinum-resistant cancer cells more responsive to PARP inhibition. On their own, PARP inhibitors shrink tumours in 25% of platinum-resistant people with a BRCA mutation, and in 5% of those without a BRCA mutation. In 2018, researchers at the Dana-Farber Cancer Institute in Boston, Massachusetts, and the NCI found that combining the PARP inhibitor olaparib with a VEGF inhibitor shrank the tumours of 20% of people with platinum-resistant disease, regardless of whether their BRCA genes were mutated5. The same drug combination is now being tested in a phase II/III trial in 680 people with several types of ovarian and fallopian-tube cancers that have either developed resistance to platinum or never responded to it.A combination approach might also make it possible for immunotherapy drugs to work against platinum-resistant ovarian cancer. Interest in immunotherapy for ovarian cancer has grown since a 2003 study found a correlation between high numbers of T cells in tumours and longer survival6. But trials of checkpoint inhibitors — drugs that encourage T cells to attack tumours — have been disappointing. In a study published in June7, for example, the multinational drug makers Merck and Pfizer reported that the checkpoint inhibitor avelumab had no impact on survival or tumour size in a phase III trial in people with platinum-resistant or unresponsive ovarian cancer.Despite these findings, Sood thinks that immunotherapy is not completely hopeless. ‘We simply need to understand the biology,’ he says. Researchers are beginning to appreciate how the proteins that are activated by DNA damage interact with the immune system. In cell-line and mouse experiments, DNA damage that accumulates in the presence of PARP inhibitors induces tumour cells to produce higher levels of the immune checkpoint protein PD-L1, which inhibits T cells. This has led to the idea that neutralizing PD-L1’s effect could give PARP inhibitors a better chance in platinum-resistant cancer.This concept was supported by a phase I/II trial funded by Waltham, Massachusetts-based Tesaro (now part of the multinational GlaxoSmithKline) in 2019, in which 62 people with recurrent platinum-resistant ovarian cancer received the checkpoint inhibitor pembrolizumab along with the PARP inhibitor niraparib. The combination shrank tumours in 19% of people without BRCA mutations compared with a 5% response rate, on average, to PARP inhibitors alone8.Several teams are also studying a triple therapy that combines VEGF inhibitors, PARP inhibitors and checkpoint inhibitors. Kohn, who has worked on some of these combination trials, says she is unimpressed by the results so far. ‘I’m not an immunotherapy nihilist,’ she says, ‘but I don’t know that we have enough information to make a scientific plan.’Checkpoint removedKohn says that data from drugs that target the cell cycle in platinum-resistant disease are more encouraging. ‘The science really makes sense there right now,’ she says.When a cell senses DNA damage, it activates a suite of proteins that bring division to a halt, giving the cell time to repair its DNA before trying to copy it. Inhibiting these proteins with a drug removes this checkpoint, and the tumour cell pushes on before any repairs can be made. This can cause cells to stall and then die when they try to divide, or to pass on DNA damage that will weaken the tumour. In high-grade serous ovarian cancer, one of these checkpoints is already damaged, so researchers are taking advantage of that biology by targeting others.Two phase II trials have tested this concept in people with platinum-resistant and unresponsive ovarian cancer. One, run by Dana-Farber and NCI researchers, tested a compound that blocks the cell-cycle control protein ATR in combination with a chemotherapy agent called gemcitabine. Of 34 participants treated in this way, half survived for 6 months without any worsening of disease, compared with 11 of the 36 individuals who received gemcitabine alone9.In another trial, published in January, researchers at the NCI and the Princess Margaret Cancer Centre found that an inhibitor of the cell-cycle control protein Wee1, used alongside gemcitabine, shrank tumours in 23% of participants compared with just 6% of those treated with gemcitabine alone. It also improved overall survival from a median of 7.2 months to 11.4 months10. ‘That was exciting,’ says Lheureux, who co-led the study. ‘It showed that if we choose a good target, platinum resistance can be overcome.’Kohn says these combinations are promising because gemcitabine works by depriving cells of nucleotides — the building blocks of DNA. So when ATR or Wee1 i
https://www.nature.com/articles/d41586-021-03718-6
