This Week in Oncology — Sep 2, 2026

Welcome to This Week in Oncology. This week we're covering 10 notable papers spanning perioperative and late-line immunotherapy, the shifting genomic and treatment landscape of colorectal cancer, and the whole-patient side of our specialty — cardiovascular risk, treatment of older adults, and the economics of biosimilars. Let's dive in.

We start with perioperative immunotherapy in lung cancer. In the Journal of Clinical Oncology, Heymach and colleagues report updated outcomes from the phase three AEGEAN trial, in which 740 patients with treatment-naive, resectable stage two to three-B non-small cell lung cancer, excluding those with EGFR or ALK alterations, were randomly assigned to four cycles of neoadjuvant platinum chemotherapy with either durvalumab or placebo, followed by twelve cycles of the same agent after surgery. At a median follow-up of about 26 months, the event-free survival benefit held up, with durvalumab cutting the risk of an event by roughly 30 percent, and disease-free survival in the resected population moved in the same direction, with about a third lower risk of recurrence. Overall survival numerically favoured durvalumab but the difference was not statistically significant and remains immature. On safety, grade three or four adverse events during the adjuvant phase occurred in about 15 percent of durvalumab patients versus about 11 percent on placebo — a real but manageable increment. The practical message is that this is confirmatory rather than new: perioperative durvalumab remains a supported option, but when you counsel a patient about twelve cycles of adjuvant therapy after an already demanding neoadjuvant course, the honest framing is a durable delay in recurrence with survival data still pending.

Colorectal cancer dominated the week, from first-line chemotherapy backbones to refractory disease to the underlying biology. Also in the Journal of Clinical Oncology, Sunakawa and colleagues report the final survival analysis of the randomised phase two DEEPER trial, which compared modified FOLFOXIRI plus cetuximab against the same triplet plus bevacizumab in RAS wild-type metastatic disease. In the overall per-protocol population there was no difference in progression-free or overall survival. The signal sits in the subgroup that matches how most of us already practise — RAS and BRAF wild-type, left-sided tumours — where median progression-free survival was about three months longer with cetuximab, roughly fifteen versus twelve months, and median overall survival was about ten months longer, fifty versus forty months, though that survival difference did not reach statistical significance. Exploratory analyses suggested longer survival with cetuximab in male patients and in those with extrahepatic disease. This is hypothesis-generating subgroup work from a phase two trial, so it should reinforce, not redefine, the anti-EGFR-for-left-sided convention.

At the other end of the disease course, Clinical Cancer Research published extended follow-up from Schlechter and colleagues on the Fc-enhanced anti-CTLA-4 antibody botensilimab combined with the anti-PD-1 antibody balstilimab in microsatellite-stable metastatic colorectal cancer without active liver metastases — a population that essentially never responds to conventional checkpoint blockade. Across 123 heavily pretreated patients, two thirds OF WHOM had three or more prior lines, the objective response rate was 21 percent, median overall survival was about 21 months, and one in three patients was alive at three years. Responses were durable, with median duration not yet reached. Diarrhoea affected about 39 percent of patients, severe in roughly one in twelve. This is a single-arm phase one-b cohort with a selected, liver-metastasis-free population, so the survival figures are flattered by selection — but a fifth OF PATIENTS responding in microsatellite-stable disease is a genuine biological signal worth tracking into randomised trials.

Underpinning all of this, Cancer Cell published an integrated clinicogenomic analysis from Manca and colleagues of more than 7,000 consecutively sequenced colorectal cancers. They mapped the timing and tropism of organ-specific metastasis, showing sequential spread in microsatellite-stable disease with brain and adrenal involvement as late events, clustering of lung, bone and brain metastases, and — importantly for the clinic — that RAS pathway activation raised the risk of spread to every metastatic site studied while WNT pathway activation was broadly protective. Combining primary tumour location with genomics can therefore begin to predict where a given patient's disease is likely to go, which is the kind of information that should eventually shape surveillance imaging.

A third thread this week is biomarker-driven precision oncology, and how unevenly it performs. Clinical Cancer Research reports a single-centre phase two basket trial from Ross and colleagues giving ado-trastuzumab emtansine to 95 patients with HER2-amplified advanced solid tumours across five cohorts. The headline overall response rate of 23 percent conceals enormous heterogeneity: salivary gland cancers responded in 58 percent of patients, with median overall survival near 29 months, while endometrial cancer responded in about a fifth of patients, lung in about one in six, and none of the seven colorectal patients responded at all. The lesson is that HER2 amplification is not a tumour-agnostic ticket — lineage still matters — but the salivary gland data are strong enough to make T-DM1 a serious consideration in that rare and otherwise option-poor population. Complementing this, Al-Sawaf and colleagues in the Journal of Clinical Oncology present a refined continuous individualised risk index for chronic lymphocytic leukaemia after fixed-duration targeted therapy, which folds in measurable residual disease at interim, end of treatment, and twelve months afterwards. Validated in the CLL14 and CLL13 trials, it clearly outperformed pretreatment indices and residual disease status alone, and stratified patients into groups with three-year progression-free survival of essentially 100 percent, 70 percent, and just 11 percent. The tool is freely accessible, and for anyone using venetoclax-obinutuzumab it offers a defensible way to tell a patient in front of you what their next three years actually look like.

And from Nature Cancer, a small early-phase study worth knowing about but not acting on: Gong and colleagues gave oral L-glutamine alongside gemcitabine and nab-paclitaxel to sixteen treatment-naive patients with advanced pancreatic cancer. The recommended phase two dose was reached at maximum glutamine dosing, tumour shrinkage occurred in nearly all participants, the response rate was 44 percent, and median overall survival was 22 months. That compares favourably with historical controls, but this is sixteen patients, single-arm, with severe treatment-related toxicity in two thirds OF PARTICIPANTS driven by the chemotherapy backbone. Randomised confirmation is essential.

Finally, three papers about the patient around the cancer. In JAMA Internal Medicine, Leong and colleagues randomised 2,487 men with prostate cancer starting androgen deprivation therapy across 55 sites in eight countries to usual care or usual care plus routine referral to an internist or cardiologist mandating a statin and a systolic blood pressure target of 130. Over a median of nearly six years, the hierarchical composite favoured the intervention — but read carefully, because the win was driven almost entirely by better cholesterol control, with a mean total cholesterol reduction of 12 milligrams per decilitre. Blood pressure differed by less than two points, and there was no difference in cardiovascular death, myocardial infarction, stroke, or heart failure. So the referral pathway improved a surrogate, not events. The pragmatic read is that if you are going to intervene, simply prescribing the statin yourself may achieve most of what a referral achieves. Alongside that, JAMA Oncology published a review by Smith and colleagues on systemic therapy in older adults with early-stage breast cancer, emphasising decision-making tools that balance competing risks — avoiding overtreatment of lower-risk disease in frail patients while not reflexively undertreating higher-risk cancers, where omission raises breast cancer mortality. And in the same journal, Liu and colleagues analysed nearly 15,000 patients receiving bevacizumab, rituximab, or trastuzumab in commercial and Medicare data. Reference product market share fell about 30 percent per year after biosimilar entry, and patients using only biosimilars incurred roughly 3,800 dollars lower monthly payer costs and about 40 dollars lower monthly out-of-pocket costs than those on the originator — a reminder that the biosimilar choice on your order set is a financial toxicity intervention.

If you only have time for one paper this week, make it the AEGEAN update in the Journal of Clinical Oncology. It is the most directly practice-relevant dataset here for anyone managing resectable non-small cell lung cancer, and it gives you the honest numbers — sustained event-free survival benefit, immature and so far non-significant overall survival — that a patient deciding about a year of perioperative immunotherapy deserves to hear.

Here are the key takeaways from this week in Oncology. Perioperative durvalumab continues to delay recurrence in resectable lung cancer, but do not promise a survival benefit yet. In first-line metastatic colorectal cancer, the anti-EGFR advantage remains confined to left-sided, RAS and BRAF wild-type disease, and even there the survival difference was not statistically significant. Botensilimab plus balstilimab produced durable responses in a fifth OF PATIENTS with microsatellite-stable colorectal cancer lacking active liver metastases — promising, but single-arm and selected. HER2 amplification predicts response very differently across tumour types, with salivary gland cancer the standout for T-DM1. And on the systems side, referring men on androgen deprivation therapy to a cardiologist improved cholesterol but not cardiovascular events, while defaulting to biosimilars measurably lowers what your patients pay.

That's your roundup for This Week in Oncology. The full transcript and references are available on the episode page in your AudioScholar library. 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 it saves you time, a quick rating in your podcast app helps other physicians find the show.

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