This week in Oncology

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

Welcome to This Week in Oncology. This week we're covering 10 notable papers spanning the shifting boundaries of immunotherapy in early-stage disease, circulating tumour DNA as a tool for prognosis and de-escalation, and new targeted and supportive treatments in harder-to-treat cancers. Let's dive in.

We start in lung cancer, where this week's evidence questions several assumptions about adding treatment to curative-intent radiotherapy. In the Journal of Clinical Oncology, Jabbour and colleagues report KEYNOTE-867, a phase three, double-blind trial of 448 patients with inoperable stage one or two non-small cell lung cancer, or who declined surgery, randomized to stereotactic body radiotherapy with either pembrolizumab or placebo [1]. Event-free survival was essentially unchanged, a median of about 31 months with pembrolizumab versus about 28 months with placebo, a difference that was not statistically significant. The harm side was clearer: severe treatment-related adverse events rose from about 4 percent to about 20 percent, and all five treatment-related deaths occurred in the pembrolizumab arm. The trial was stopped early on benefit-risk grounds. This is a firm negative result from a large blinded trial, and it does not support adding checkpoint blockade to stereotactic radiotherapy in this population. A companion question, whether stereotactic radiotherapy can stand in for surgery in patients fit for either, is addressed by Kagenaar and colleagues, also in the Journal of Clinical Oncology [2]. Because randomized trials have struggled to recruit, they emulated a target trial using English national cancer registry data on more than twelve thousand patients with node-negative tumours, using an instrumental variable approach to reduce confounding by indication. After adjustment, three-year all-cause mortality was roughly nine percentage points higher after stereotactic ablative radiotherapy, about 30 percent versus about 22 percent after surgery. The difference in lung cancer-specific mortality pointed the same way but did not reach significance, and among patients aged 80 and over the gap was smaller and not significant. The authors conclude that surgery should remain the preferred option where feasible, while radiotherapy remains a curative-intent alternative, particularly for older people. As observational evidence, however sophisticated, it cannot fully exclude residual confounding, and ongoing randomized trials will matter.

Further along the disease course, JAMA Oncology carries the R-ALPS trial from Ji and colleagues, a phase three study at 42 centres in China testing consolidation after chemoradiotherapy for unresectable stage three disease [3]. The anti-PD-L1 antibody benmelstobart, with or without the multi-target kinase inhibitor anlotinib, clearly improved progression-free survival over placebo, roughly halving the risk of progression, with median progression-free survival around 15 months for the combination and about 10 months for benmelstobart alone, against about 4 months on placebo. But in the pooled head-to-head comparison, adding anlotinib did not significantly improve progression-free survival over benmelstobart alone, and it raised fatal treatment-related events to about 7 percent compared with under 2 percent in the other arms. Read alongside KEYNOTE-867, the theme is that intensifying immunotherapy-based regimens in curative settings carries real toxicity costs that efficacy gains do not always justify. Finally for lung, ASCO has issued an update to its living guideline on therapy for stage four non-small cell lung cancer with driver alterations, led by Reuss, again in the Journal of Clinical Oncology [4]. The abstract describes the living-guideline process rather than specific recommendations, so the details sit in the full document.

Our second theme is immunotherapy reaching into gastrointestinal cancers once thought largely resistant to it. In the Journal of Clinical Oncology, Bai and colleagues randomized 392 patients with borderline resectable or locally advanced pancreatic cancer to modified FOLFIRINOX with or without the PD-1 inhibitor toripalimab [5]. The combination significantly improved event-free survival, cutting the risk of an event by about a quarter, though the median gain was modest, about a month and a half. Overall survival trended in the same direction, roughly 21 versus 17 months, but this difference was not statistically significant. Response rates rose from about a fifth of patients to about three in ten, resection rates were similar, and severe toxicity and surgical complications were not meaningfully increased. Prespecified subgroups suggested the benefit was concentrated in borderline resectable rather than locally advanced disease. This is an open-label phase two trial from a single country, so it signals promise rather than settling practice, and confirmation would be needed before guidelines move. In colon cancer, Nature Medicine publishes the UNICORN trial from Ghelardi and colleagues, testing short-course preoperative botensilimab, an Fc-enhanced anti-CTLA-4 antibody, alone or with the anti-PD-1 antibody balstilimab, in 56 patients with resectable locally advanced disease [6]. Mismatch repair-deficient tumours treated with the combination all achieved a major pathological response. More striking, in mismatch repair-proficient tumours, long considered immunologically cold, the combination produced major responses in about a third of patients, whereas botensilimab alone produced none. Immune-related adverse events affected just over half of patients on the combination, mostly flu-like symptoms and diarrhoea. With 14 patients per cohort, this is hypothesis-generating, but it reopens the question of neoadjuvant immunotherapy for the majority of colon cancers that are proficient.

Our third theme is circulating tumour DNA moving from research curiosity toward a practical tool for prognosis and treatment tailoring. In Nature Medicine, Pérez-García and colleagues report five-year results from PHERGain, a non-comparative phase two trial in HER2-positive early breast cancer that used PET response after two cycles and pathological complete response to decide who could skip chemotherapy [7]. Across the de-escalation arm, five-year invasive disease-free survival was about 90 percent, and among the 86 PET responders who reached complete response without chemotherapy it was about 92 percent, with only one distant recurrence. The authors are careful to frame that subgroup estimate as descriptive rather than proof that omitting chemotherapy is safe. Detectable circulating tumour DNA predicted worse outcomes, and clearance tracked with higher complete response rates. In The Lancet Oncology, Yip and colleagues present an exploratory biomarker analysis of TALAPRO-2, the phase three trial of talazoparib plus enzalutamide in metastatic castration-resistant prostate cancer [8]. Homologous recombination repair status from liquid biopsy agreed with tissue testing in about 84 percent of evaluable patients. Detectable baseline tumour DNA was strongly prognostic in both arms, roughly two and a half times the risk of progression with the combination and about four times with enzalutamide alone. The colon cancer data in UNICORN, where pre-surgical clearance of tumour DNA correlated with response and freedom from recurrence, point the same way [6]. Across three tumour types, these remain exploratory or secondary analyses, so they support the biomarker's prognostic validity but not yet its use to direct therapy.

Our last theme brings two papers, both in the Journal of Clinical Oncology, addressing areas with few options. McKean and colleagues report the phase two expansion of OptimUM-01, combining darovasertib, a first-in-class oral protein kinase C inhibitor, with the MET inhibitor crizotinib in 44 previously untreated patients with metastatic uveal melanoma [9]. About a third of patients responded, roughly nine in ten had disease control at eight weeks, and median overall survival was about 21 months. Gastrointestinal effects were nearly universal, and pulmonary embolism was the most common severe event at about 9 percent, though only three patients stopped for toxicity. It is single-arm, so it cannot establish superiority, but it offers a biologically targeted option independent of HLA type. And Frederiksen and colleagues tested methylphenidate for fatigue in 152 patients with hematologic cancers using a double-blind, placebo-controlled crossover design [10]. Fatigue scores improved by about five points more on methylphenidate than on placebo, a finding echoed by other fatigue measures. Adverse events were mild but more frequent on the drug, reported in about 57 percent of treatment periods versus 40 percent on placebo. Patients were off chemotherapy for at least eight weeks, which limits generalization to those on active treatment.

If you only have time for one paper this week, make it KEYNOTE-867 [1]. It is a large, blinded, definitively negative trial that closes the question of adding pembrolizumab to stereotactic radiotherapy in early-stage inoperable lung cancer, with a toxicity and mortality signal that reframes how intensification in frail patients is weighed.

Here is what this week's evidence adds up to in Oncology. First, in early-stage lung cancer, phase three data show no event-free survival benefit and clear harm from adding pembrolizumab to stereotactic radiotherapy, while large registry data suggest surgery carries lower three-year mortality than radiotherapy in patients fit for both, with randomized confirmation still pending. Second, consolidation immunotherapy after chemoradiotherapy is reinforced by R-ALPS, but layering on anlotinib added deaths without a significant efficacy gain. Third, immunotherapy is edging into pancreatic and proficient colon cancer, with a significant event-free survival gain but no proven overall survival benefit in pancreatic disease, and only small phase two cohorts in colon cancer. Fourth, circulating tumour DNA is consistently prognostic across breast, prostate, and colon cancer, but evidence that it should guide treatment decisions remains exploratory. And finally, methylphenidate showed a modest, statistically significant fatigue benefit in hematologic cancer patients off chemotherapy, at the cost of more side effects.

That's your roundup for This Week in Oncology. The full transcript and references are available on the episode page. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.