NEW YORK – While targeted therapies can dramatically shrink tumors for patients, the benefits are often temporary and patients inevitably acquire mutations that help cancer cells survive, requiring a different treatment approach.
Typically, drugmakers and researchers try to overcome this resistance by either replacing the original therapy causing resistance with a new treatment that blocks the emerging resistance mutation or by inhibiting both the original target mutation and the new acquired resistance mutation through combination treatments.
However, it's hard to anticipate the mechanism that patients' cancer cells will exploit for survival, and combination therapies can add to the burden of unwanted toxicities for already sick cancer patients. To address these challenges, researchers are looking for more universal signals of resistance that, if blocked, can resensitize tumors to multiple treatments and are leveraging new tools like circulating tumor DNA (ctDNA) tests to monitor for acquired resistance mutations.
Robert Clarke, executive director of the Hormel Institute at the University of Minnesota who researches chemo and endocrine resistance in hormone receptor (HR)-positive breast cancer, compared the process of identifying and addressing treatment resistance to a jigsaw puzzle, "except we don't have the picture of what we're supposed to solve for and some of the puzzle pieces are blank."
"We're trying to put this puzzle together and make sense of it without the full picture," he said.
Clarke suggested that approaching resistance from a single mutation or overexpressed gene is too simple and that taking a systems biology perspective may be what's needed to fully eradicate cancer resistance. The heterogeneous nature of many tumors means that cancer cells have many ways to potentially escape treatment, he said, and researchers should take that same wide approach when looking at resistance.
Research about treatment resistance mechanisms used to bypass KRAS G12C inhibitors, for example, has involved more precise sequencing methods, such as single-cell sequencing to identify all the alterations that can play a role in acquired resistance to inform new drugs and drug combinations.
"Despite the meaningful improvements in disease control and tumor response with these [first-generation] inhibitors of KRAS, they have some limitations on the durability of those responses," said Scott Kopetz, deputy chair for translational research in the department of gastrointestinal medical oncology at the University of Texas MD Anderson Cancer Center. "Like many of our chemotherapies and other cancer treatments, resistance ultimately develops in these patients."
Expecting resistance
For decades KRAS was considered an "undruggable target." And while the first KRAS inhibitors to come to market, Amgen's Lumakras (sotorasib) and Bristol Myers Squibb's Krazati (adagrasib), represent a major scientific advance for patients, drugmakers didn't rest on their laurels, as they knew they'd have to tackle primary and acquired resistance to these KRAS G12C inhibitors. In the few years Lumakras and Krazati have been on the market, drugmakers have begun testing scores of combination approaches trying to address resistant tumors.
Resistance is such a common occurrence with molecularly targeted medicine that the research community knew this would be a problem with the first-generation KRAS inhibitors, according to Kopetz. Other targeted therapies, like EGFR inhibitors, have shown similar patterns of resistance, where cancer cells use other genetic pathways to escape the targeted inhibition.
"It was no surprise when we saw the patterns of resistance in KRAS," Kopetz said. "It's very similar to what we see when using EGFR inhibitors alone or using BRAF inhibitor combinations, so it was expected and anticipated in this setting."
In colorectal cancer trials, for example, patients developed both primary resistance by failing to respond to KRAS G12C inhibitors from the start and acquired resistance after initially responding to them. While Krazati and Lumakras both received monotherapy approvals in KRAS G12C-mutant lung cancer, these drugs initially showed poor monotherapy performance in colorectal cancer, spurring their sponsors to quickly pair them with EGFR monoclonal antibodies, and these combination approaches eventually gained regulatory approval.
In a Phase I/II trial that compared single-agent Krazati to Krazati combined with Eli Lilly's EGFR inhibitor Erbitux (cetuximab), for example, 19 percent of those on the monotherapy responded to treatment, compared to 46 percent in the combination arm. The combination arm also had better median progression-free survival, 6.9 months, compared to 5.6 months in the monotherapy arm.
"In the KRYSTAL-10 study, by combining [Krazati] with an anti-EGFR therapy in patients with previously treated colorectal cancer, we are hoping to delay when acquired resistance may impact treatment," a BMS spokesperson said.
Lumakras' efficacy was similarly impacted by primary resistance when given as a single-agent in KRAS G12C-mutant colorectal cancer patients. Amgen also pivoted to a combination approach with an EGFR inhibitor after treatment with single-agent Lumakras showed only a 9.7 response rate and median progression-free survival of 4 months in a Phase II colorectal cancer study. Adding Amgen's EGFR monoclonal antibody Vectibix (panitumumab) to Lumakras improved these outcomes and demonstrated a 26 percent response rate and 5.6 months median progression-free survival.
Kopetz noted that many colorectal cancer patients also experience acquired resistance to KRAS G12C inhibitors, as after a period of initial benefit, their tumors start to progress again. "The strategies have been somewhat limited in how we target resistance mutations just because there's a wide number of them that can appear [in response to KRAS inhibition]," Kopetz said.
Studies of acquired resistance to KRAS G12C inhibitors have shown that these tumors use escape mechanisms in the RAS/MAPK pathway, such as alterations in BRAF, NRAS, MEK, ERK, and EGFR, to resist KRAS G12C inhibition. Kopetz also noted that cancer cells can exploit resistance mechanisms within KRAS itself, such as KRAS amplifications, and in other genetic pathways such as YAP/TAZ-TEAD and PI3K/AKT.
One promising approach for addressing KRAS G12C inhibitor resistance is to combine it with newer pan-RAS therapies, such as Revolution Medicines' daraxonrasib and Erasca's ERAS-0015. Along with expanding the patient population to include other KRAS variants, like G12D or G12V, these drugs may also help address on-target resistance mutations. "If we have a drug targeting a specific KRAS mutation, and then we have secondary mutations in KRAS that appear, can we take that targeted inhibitor and combine it with a pan-RAS inhibitor?" Kopetz said. "The hope is that approach may be one way to extend the duration to benefit and potentially address secondary mutations in RAS, as well as amplification events."
In addition to trying to use treatment strategies to target and shut down the various pathways that cancer cells are using to survive KRAS inhibition, researchers are also testing out combinations with chemotherapy and immunotherapy. "We want to extend the benefit for patients who do respond and lengthen the duration of time they are on treatment; that's where we are addressing acquired resistance," he said. "Also, not all the patients respond well on first exposure, so the question there is: how do you broaden the population of responding patients?"
BMS, for example, is conducting several studies of Krazati combined with immunotherapy and chemo to address potential resistance, according to the company spokesperson, including the Phase II KRYSTAL-4 and KRYSTAL-7 studies in KRAS G12C-mutant NSCLC.
Experts like Kopetz are particularly hopeful that the cross talk between immunotherapy and KRAS inhibitors may help both treatments work better. "If you inhibit KRAS, you may prime the tumors to being more responsive to immunotherapy," he said. "The combination there may get more patients to benefit beyond what we're seeing with the KRAS inhibition alone."
So far, that hypothesis appears to be panning out in clinical trials. In updated results from the KRYSTAL-7 trial researchers presented earlier this year, patients with PD-L1-positive KRAS-mutated NSCLC who received Krazati plus Merck's checkpoint inhibitor Keytruda (pembrolizumab) in the first-line setting had a median progression-free survival of more than 27 months, a significant improvement compared to the median progression-free survival of 6.5 months that patients achieved with single-agent Krazati in later lines.
BMS is also focused on conducting biomarker assessments and identifying genomic signatures to better predict treatment benefit in its trials, according to the spokesperson. Moreover, these ongoing combination trials will help inform development of BMS's other earlier targeted therapies, the spokesperson added.
A common marker
Kairos Pharma, meanwhile, has identified CD105 as a biomarker of resistance that appears to occur across multiple tumors and treatment types. Its lead candidate, ENV-105, targets CD105 as a way to reverse acquired resistance to EGFR inhibitors and hormone therapy.
When cancer cells are under stress and trying to survive attack from a targeted therapy, for example, CD105 is upregulated and promotes bone morphogenetic protein (BMP) signaling, which plays a role in cell proliferation, differentiation, and apoptosis. By inhibiting CD105 and BMP signaling, Kairos' ENV-105 is designed to switch off the stress response and the associated cell survival mechanisms, said Kairos CSO Neil Bhowmick.
"The preclinical studies showed that if we give EGFR inhibitors, CD105 goes on, and it allows for a resistance mechanism," Bhowmick said. "Then if you block it, the EGFR inhibitor works more effectively."
Currently, Kairos is studying ENV-105 in two early-stage trials involving patients with EGFR-mutant NSCLC who are resistant to EGFR inhibitors and patients with metastatic castration-resistant prostate cancer who are resistant to anti-androgen therapy. In both studies, patients are receiving ENV-105 alongside the treatment they've developed resistance to, AstraZeneca's EGFR inhibitor Tagrisso (osimertinib) for NSCLC and Janssen's androgen receptor inhibitor Erleada (apalutamide) for prostate cancer.
In the lung cancer trial, Kairos is enrolling NSCLC patients who have developed complete resistance to the EGFR inhibitor and have progressed on treatment, as well as those who've had a partial response. Bhowmick said many patients experience treatment resistance within 18 months of receiving an EGFR inhibitor, and few experience a complete response to this treatment.
"We're seeing if we can improve a [partial] response by doing the combination therapy [with ENV-105]," he said. "Even if 80 percent of patients initially respond, it may not be a full response, so there's a significant number of EGFR-mutated lung cancer patients who would potentially benefit with this combination therapy."
Similarly in prostate cancer, patients often progress on treatment with hormone therapy. In the first-line setting, progression and treatment resistance often occur after about 18 months, and in the second-line setting, resistance can occur as early as two months on treatment with hormone therapy.
In a second Phase II study, Kairos is evaluating ENV-105 alongside anti-androgen therapy in prostate cancer patients who have progressed on prior anti-androgen treatments or who had rising prostate-specific antigen (PSA) levels while on these treatments. "First-line hormone therapy [in prostate cancer] works for about a year and a half and second-line hormone therapy for about 2.5 months," Bhowmick said. "We're seeing if we can improve second-line hormone therapy over that 2.5 months with the addition of ENV-105."
Cancer cells also become stressed and activate the CD105 pathway when attacked by hormone therapy, such as in prostate cancer treatment. Similar to how ENV-105 helps address resistance to EGFR inhibitors, it can also address resistance to hormone therapy by targeting the CD105 pathway.
The firm has also started exploring whether this CD105 survival pathway is activated in response to radiation therapy in head and neck cancer, finding that this treatment also induces hypoxia in cancer cells and activates CD105, he said. "So far, the few areas of drug resistance that we've looked at, it seems the common theme has been this survival pathway," Bhowmick added.
While Kairos has not yet reported results from either of these ENV-105 trials, which only began in the past year, the firm expects a safety and efficacy data readout from the prostate cancer study in the first half of 2025. In a recent preclinical publication, the firm reported that ENV-105 restored sensitivity to Tagrisso within resistant EGFR-mutant NSCLC cell lines and mouse models.
"[Our team] has discovered a central mechanism of resistance that appears to impact a host of diseases and allows the drugs to work again," said Kairos CEO John Yu. "I see this therapy as a means of making a lot of therapies work again when they fail. There's a lot of drugs that work great at first, but almost all of them fail, and to be able to harness a resistance mechanism and to reverse it is meeting a vast unmet medical need."
Readying for resistance with ctDNA
The emergence of ctDNA testing has provided doctors and drugmakers alike with a tool for flagging early signs of treatment resistance, so they can start planning the most optimal treatments for patients before they show signs of cancer progression on imaging scans. In the Phase III SERENA-6 trial, for example, AstraZeneca is exploring whether hormone receptor (HR)-positive HER2-negative breast cancer patients who acquire a mutation associated with aromatase inhibitor resistance during the study can be switched to its next-generation selective estrogen receptor degrader camizestrant and have better outcomes.
The trial investigators are using ctDNA to monitor patients for an acquired ESR1 mutation, which is known to drive resistance to aromatase inhibitors and endocrine therapy. Once this mutation is identified in a patient in the study, they are randomized to either camizestrant plus a CDK4/6 inhibitor and placebo or to remain on the aromatase inhibitor with a CDK4/6 inhibitor and placebo.
Teresa Klinowska, global product leader in late-stage oncology R&D at AstraZeneca, said that when patients with HR-positive HER2-negative breast cancer start standard frontline treatment with endocrine therapy and a CDK4/6 inhibitor, only around 5 percent are likely to have an ESR1 mutation.
"Over time, particularly on aromatase inhibitor therapy in combination with the CDK4/6 inhibitor, you get the emergence of mutations in the ESR1 gene," she explained. By the time patients progress on this first-line treatment in this setting, ESR1 mutation occurrence has typically increased to about 40 percent, she estimated. Researchers' goal within the SERENA-6 trial is to try to catch patients with ESR1 resistance mutations before their cancer progresses.
"The thinking was instead of waiting for disease to progress into the second-line setting, which is a much more traditional way of changing patients' therapy, we were taking advantage of this new technology," Klinowska said. "We were looking at circulating tumor DNA to find the emerging resistance at the earliest possible moment and get ahead of disease progression by optimizing the endocrine therapy with camizestrant while continuing the CDK4/6 inhibitor."
Interim data from SERENA-6, which will be reported at an upcoming medical meeting, showed that patients on camizestrant combination therapy had a statistically significant and clinically meaningful improvement in median progression-free survival compared to patients on standard-of-care therapy with an aromatase inhibitor and a CDK4/6 inhibitor.
Currently, Menarini Group's ESR1-targeting SERD Orserdu (elacestrant) is approved in the US in the second-line setting for HR-positive, HER2-negative breast cancer patients that progressed on first-line endocrine therapy and have an ESR1 mutation. However, AstraZeneca is aiming for a potential first-line approval for camizestrant with the SERENA-6 trial design, using ctDNA to monitor for the resistance mutation before progression.
"We hope to keep [patients] in that first-line setting for as long as possible," Klinowska said. "When they then progress into second line, they're on a treatment trajectory with a higher toxicity burden and a lower efficacy return for the next lines of therapy."
Financial, treatment toxicity are barriers
For patients, these resistance-combating strategies are not without risks, as they could lead to heightened toxicity from combining multiple targeted therapies. The KRAS combination approaches with immunotherapy and chemo, for example, demonstrated relatively high rates of grade 3 or higher adverse events. Krazati plus Keytruda led to 69 percent of patients in a trial experiencing a grade 3 adverse event, while in another Phase I study, half of the patients experienced a grade 3 event on Krazati.
Kopetz said that in the case of KRAS and EGFR inhibitor combinations, the benefits come with little increase in toxicity, making the choice easier on treating physicians.
For example, one colorectal cancer trial of Krazati and Erbitux reported that grade 3 or 4 treatment-related adverse events were lower in the combination arm, 16 percent, than in the Krazati monotherapy arm, 34 percent.
Kairos is also managing the toxicity profile of its combination approaches with ENV-105. While Bhowmick said ENV-105 has a manageable profile, with very little toxicity reported, the firm also must account for any toxicity that may come with its combination partners, like Tagrisso and anti-androgen therapy in its current trials.
"Because we're not doing single-agent treatment, we're going to take the toxicity of hormone therapy along with the toxicity of our own drug," Bhowmick said. "The combination thus far hasn't been bad, and we've specifically picked apalutamide [as the anti-androgen therapy partner] because it doesn't have significant toxicity on its own."
Meanwhile, the SERENA-6 trial of camizestrant presents an opportunity for ctDNA monitoring for resistance mutations in breast cancer. While ctDNA is sometimes used in clinical practice to test for baseline tumor mutations, right now, it is not commonly used in practice as a resistance monitoring tool, Kopetz said.
Klinowska said integrating ctDNA into the SERENA-6 trial has been seamless, since patients come into the clinic for regular blood draws and scans anyway while on the study. The challenge, however, has been shifting how physicians think about resistance monitoring and explaining to patients about changing treatment, she said.
"The logistical challenges of the trial were explaining the design to the investigators and getting them to think about monitoring for mutations during first-line [treatment], instead of waiting for progression to second-line [treatment] and explaining the benefit to patients and physicians of intervening early rather than later, after progression to optimize their outcomes," she said.
These resistance strategies also raise the issue of balancing benefits against toxicity, both financial and clinical. Combination therapies cost more than a single-agent treatment would, and new drugs usually have higher costs than old treatments that may have generic options available. When Krazati was first approved, its list price was about $19,000 for a 30-day supply of the drug, according to reports. Adding an EGFR inhibitor on top of that can nearly double the cost before insurance, as Erbitux, for instance, has a list price of $16,242.60 per month.
Kopetz noted that toxicity and cost may factor into the choice for KRAS inhibitor combos, but that doctors "would prioritize combinations that meaningfully improve response rates to justify any additional toxicity and cost," he said, adding that Krazati plus Erbitux had a response rate that was nearly double that of the single-agent response rate, 34 percent versus 19 percent, respectively.
Clarke took a more skeptical stance, arguing that new drugs addressing resistance in breast cancer should meaningfully improve overall survival. While many new breast cancer drugs like CDK4/6 inhibitors or HER2 inhibitors have improved progression-free or disease recurrence-free survival, overall survival has been more of a challenge, he said.
"Patients that recur while on treatment, that's a classic drug resistance phenotype," Clarke said, adding that resistance can also cause cancers to return many years later. "[For patients who experience long remission], the cancer coming back 15 or 20 years later is resistance, too, because otherwise they'd be cured."
He is still committed to personalized medicine as the key to addressing breast cancer resistance, however. Figuring out drug combinations that work within certain molecular subgroups, such as the treatments for ESR1-mutant breast cancer patients, or developing tests to monitor and predict patient outcomes on certain treatments, like the ctDNA monitoring, are critical to this area of research.
"It's a fascinating, rapidly changing area of research where the potential is great and the rewards are off the charts," Clarke said. "[HR-positive breast cancer] is overtaking lung cancer as a cause of mortality for women. If we could find treatments that were cheap and easy [to deliver] so that we solve the problem in a way that's accessible for everyone, that would be just wonderful."