In 2020, pembrolizumab became a first-line treatment for certain patients with metastatic colorectal cancer (CRC).
The FDA approval essentially moved pembrolizumab from a later-line option to the frontline setting, marking an important change for patients with microsatellite instability-high or mismatch repair-deficient disease.
Thanks to the shift, pembrolizumab has become the first and only drug some patients ever take, said Mark Lewis, MD, a gastrointestinal oncologist at Intermountain Healthcare in Murray, Utah. “The slight nuance there is, what do you do if they progress on pembrolizumab? That’s a harder issue,” Lewis said.
Oncologists typically give a “stronger” immunotherapy option, such as ipilimumab plus nivolumab, Lewis said, “but really, this is a bit of an evidence-free zone.”
The fast pace of drug advances in oncology is something to celebrate. But those advances can also create a new problem: as FDA approvals increasingly move cancer drugs from later‑ to first‑line indications, oncologists are facing an evidence gap: What to prescribe when the first-line option fails and there’s no clear next option?
“It sounds crazy to complain that we’ve made improvements so quickly,” said Kathy D. Miller, MD, a breast oncologist at Indiana University Health Simon Cancer Center in Indianapolis. But “it’s simply acknowledging that we have a new group of patients for whom there are important management questions.”
The problem may be more common in some cancer types and drug classes than others, but when it does arise, the underlying challenge is the same: treatment sequences are changing faster than the evidence needed to determine what should come next. And oncologists may be operating in a space with no guideline‑supported or FDA-approved options.
This is an “uncomfortable” place for oncologists and patients, Miller said. “I can guarantee you, it will lead to significant heterogeneity in how those patients are managed.”
Faster Approvals, More Uncertainty
Oncologists have always grappled with uncertainty. Even when there’s supporting evidence, the data can’t capture every nuance, said Mark G. Kris, MD, a thoracic medical oncologist at Memorial Sloan Kettering Cancer Center in New York City.
This uncertainty has only intensified as drug approvals have multiplied.
“As we move forward, there are more nuances,” Kris said. And “there are more decisions.”
Today, a new drug can shift lines within the span of months. Take the TKI zongertinib in non-small cell lung cancer (NSCLC).
In August 2025, zongertinib was approved as a later-line therapy for unresectable or metastatic non-squamous NSCLC. Just 6 months later, it was approved in the first line. Sevabertinib, another HER2-targeted TKI, was recently approved in September for first-line use in locally advanced or metastatic non-squamous NSCLC in patients with HER2-tyrosine kinase domain activating mutations, just 10 months after earning accelerated approval for later-line use.
The shifts are already leading to questions about what to do if a patient progresses on the first-line option.
For Biagio Ricciuti, MD, PhD, a thoracic medical oncologist at the Dana-Farber Cancer Institute in Boston, the treatment trajectory could follow one of two paths.
Ricciuti often prefers zongertinib in the first-line setting, given its preferable safety profile, but if patients progress, he may opt for sevabertinib in the second-line setting. Emerging data suggest sevabertinib retains sensitivity to secondary HER2 mutations that can cause resistance to zongertinib.
However, for patients without any “clearly actionable resistance mutation,” Ricciuti said, he might try the antibody-drug conjugate (ADC) trastuzumab deruxtecan (T-DXd), which is approved for previously treated HER2-mutant NSCLC. There are also multiple chemotherapy regimens approved in the second-line setting, though none has been tested following progression on zongertinib or sevabertinib.
So is it better to give another TKI, an ADC, or chemotherapy after zongertinib or sevabertinib fails?
It’s unclear. But “we are definitely spoiled for choices now,” said Melissa L. Johnson, MD, director of lung cancer research and chief scientific officer at Sarah Cannon Research Institute in Nashville, Tennessee.
A T-DXd Data Gap
The data gap problem can become more complicated when a drug moves through multiple settings and sequences.
This challenge has arisen with T‑DXd in breast cancer, after the ADC recently received several FDA approvals in rapid succession that moved it earlier in the treatment pathway.
T-DXd was initially approved for later-line use in previously treated metastatic HER2-positive breast cancer in 2019 and HER2-low advanced breast cancer in 2022. In December 2025, T-DXd plus pertuzumab was approved as a first-line treatment for metastatic HER2-positive breast cancer. After 5 months, T-DXd followed by taxane, trastuzumab, and pertuzumab moved into the neoadjuvant setting for HER2-positive stage II or III breast cancer. Separately, T-DXd was approved as adjuvant treatment for patients with residual invasive disease after neoadjuvant trastuzumab, with or without pertuzumab, and taxane-based therapy.
Most patients will now receive T‑DXd as part of their neoadjuvant therapy, Miller said, but the adjuvant approval “has minimal relevance.” If patients don’t have a pathologic complete response following neoadjuvant T-DXd, “it makes absolutely no sense to give them more of the same drug after surgery.”
So what should come next for patients who have residual disease after neoadjuvant T-DXd?
It’s not clear, Miller said.
Miller pointed to potential options for ongoing HER2-positive disease, including a different chemotherapy regimen, the ADC ado-trastuzumab emtansine (T-DM1), or a HER2-targeted TKI, such as tucatinib or neratinib. But her decision would depend on the patient’s level of residual disease, comorbidities, and how well they tolerated their neoadjuvant therapy.
“I probably will not be consistent in what I do, because all of those factors will play a role in the decision, and because there isn’t any data to guide us,” Miller said.
Pembro: A Sequencing Black Box
For drugs like pembrolizumab that have undergone many approvals in multiple cancer types, the sequencing issues aren’t confined to one disease or drug class.
Beyond metastatic CRC, pembrolizumab’s rapid approvals in multiple cancer types have created data gaps in triple-negative breast cancer (TNBC).
In late 2020, pembrolizumab was approved for use with chemotherapy for PD-L1-positive metastatic TNBC in patients whose tumors have a combined positive score of at least 10. In July 2021, pembrolizumab was also approved in the neoadjuvant setting alongside chemotherapy for patients with high-risk, early-stage TNBC, followed by pembrolizumab monotherapy after surgery to help prevent recurrence.
“If you had pembrolizumab in the curative setting, and now your disease has progressed, there is no data on whether retreatment, regardless of what the disease-free interval might be, has any benefit,” Miller said. Some oncologists won’t treat with pembrolizumab again if the patient progressed within a “relatively short” time, Miller said.
But the cutoff points can vary from one clinician to the next, and “I don’t envision we’re ever going to do a clinical trial that addresses that question,” Miller said.
Emil Lou, MD, PhD, faces his own dilemma in patients with gastric cancer. Pembrolizumab jumped to the first-line setting for locally advanced unresectable or metastatic disease in HER2-positive gastric or gastroesophageal junction adenocarcinoma where tumors express PD-L1. Patients receive pembrolizumab in combination with trastuzumab, fluoropyrimidine- and platinum-containing chemotherapy.
Although this patient population has an approved second-line option, that regimen was studied and approved before pembrolizumab moved into the first-line setting. As a result, evidence to guide what comes next after progression on first-line pembrolizumab-based therapy may not be clear cut.
Lou said he might try combining paclitaxel chemotherapy with ramucirumab targeted therapy. Some oncologists may also continue pembrolizumab and switch out the chemotherapy to potentially “extract more benefit, but that’s an example of an area where more data are needed,” said Lou, a professor of medicine in the Division of Hematology, Oncology, and Transplantation at the University of Minnesota in Minneapolis.
“When progression occurs, we don’t have a way to assess with certainty” if one drug failed or if the whole combination did, Lou explained. And that also means “we don’t have exact evidence to tell us what sequence of therapies might be superior.”
“It’s a bit of a Wild West out there at this point,” he said.
Insurance Uncertainties
These uncertainties may extend beyond clinical decision-making. When a treatment is being used in a sequence that has not been directly studied, it may also be unclear how insurers will interpret coverage for the next therapy.
T-DM1, for example, is approved for adjuvant in HER2-positive early breast cancer in patients with residual disease after neoadjuvant trastuzumab and a taxane, “but none of those patients had a T-DXd-containing regimen,” Miller said.
Given that the FDA-approved second-line options are not based on data from patients treated with T-DXd, “I have no idea what insurance will do for somebody who had a trastuzumab-containing neoadjuvant regimen and has residual disease,” Miller said.
When there are multiple later-line options, but a lack of evidence to support one over another, payers may deny coverage.
In metastatic CRC, for example, there are three “basically equivalent” later-line options — trifluridine/tipiracil plus bevacizumab, regorafenib, or fruquintinib, Lewis said.
Regardless of his own preference, the insurer might say, “‘no, you’ve got to do this one first,’ even if there’s zero evidence that that’s what you’re supposed to do,” Lewis said. “The final stakeholder in all of these line sequencing questions are the payers.”
New Data Gaps
The growing number of orphan later-line spaces even has some oncologists anticipating potential future data gaps.
Daraxonrasib, the first-in-class pan-RAS inhibitor recently approved for previously treated metastatic pancreatic adenocarcinoma, is now being tested in a phase 3 trial as a first-line treatment, alone or with chemotherapy. But if daraxonrasib moves into the first-line setting, the treatment sequence could again change faster than the evidence needed to guide what comes next when the disease progresses.
Across these examples, the central challenge is often not a lack of treatment options, it’s a lack of evidence for what should come next. And as more drugs move into earlier lines, oncologists may increasingly have to make decisions about the best next treatment before the evidence and guidelines can catch up.
In the absence of clear evidence, the answer often comes down to individualizing treatment, Miller said. She considers the available data, weighs tolerability and toxicity, and spends considerable time talking through the options with patients.
Ultimately, she said, “these difficult decisions require long and honest conversations with the patient.”
How a patient was affected by a previous therapy can be “a pretty good tiebreaker,” Lou said. There might be two good options after progression on a certain chemotherapy, but if the patient had excessive fatigue on chemotherapy, he’ll suggest something with a different side effect profile.
Those kinds of decisions are not going away. Along with the dizzying pace of approved new drug targets and targeted therapies, Lou said, the rise of genomic profiling means more patients will become eligible for a growing list of drugs.
“You could call it a complicating factor,” Lou said. “But it’s also good.”
The sources cited in this article had no relevant disclosures.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.