This Week in Oncology — Aug 19, 2026

Welcome to This Week in Oncology. This week we're covering 10 notable papers spanning long-term outcomes and biomarker refinement in hormone-driven cancers, resistance and new drug platforms in oncogene-driven lung cancer, chemotherapy-free strategies in lymphoma, and a cluster of papers about drug safety and how we read so-called negative trials. Let's dive in.

We'll start with hormone-driven disease, where two papers in Annals of Oncology push in the same direction: longer follow-up and better biomarkers change who we treat. The ENZAMET trial, reported by Zhang and colleagues, now has eight years of follow-up in metastatic hormone-sensitive prostate cancer, comparing enzalutamide with a first-generation non-steroidal anti-androgen on top of testosterone suppression [4]. The survival advantage has held. Median overall survival was just under eight years with enzalutamide versus under six years with the older anti-androgen, and at eight years half of the enzalutamide group were alive compared with about forty percent of the control arm — roughly a quarter reduction in the risk of death. Clinical progression-free survival at eight years was more than doubled, at about forty-three percent versus twenty percent. What I find most useful clinically is the granularity on late toxicity and durability. Of the 622 deaths, most were from prostate cancer, and those were clearly fewer with enzalutamide, while deaths from other causes were essentially identical between arms — so there is no signal that we are trading cancer deaths for cardiac or neurologic ones. Grade three or higher cardiac and nervous system events were the same rate in both groups; falls were more frequent with enzalutamide, but at a low absolute rate. Median treatment duration was nearly five years with enzalutamide versus under two years with the anti-androgen, and among the third of participants still on drug at cut-off, almost ninety percent were holding the full 160 milligram dose. That is a strong argument against reflexive dose reduction for fatigue in patients who are tolerating therapy.

The second paper in this theme may change a conversation you have every week in the breast clinic. Kalinsky and colleagues went back to RxPONDER, the trial in hormone receptor-positive, HER2-negative, node-positive breast cancer with a recurrence score of 25 or less, where chemotherapy benefit was confined to premenopausal women [5]. They measured pretreatment ovarian reserve hormones in over fifteen hundred participants under age 55. Estradiol, progesterone, luteinising hormone and follicle-stimulating hormone told them nothing about chemotherapy benefit. Anti-Müllerian hormone did. Among the sixty-four percent OF WOMEN with anti-Müllerian hormone at or above ten picograms per millilitre — that is, preserved ovarian reserve — adding chemotherapy to endocrine therapy cut invasive disease-free survival events by more than half. Among the thirty-six percent OF WOMEN with low ovarian reserve, there was no benefit at all; the point estimate actually favoured endocrine therapy alone. Distant relapse-free survival followed the same pattern, and an ultrasensitive inhibin B assay was also predictive. Critically, anti-Müllerian hormone outperformed menopausal status, age, and every other hormone as a marker of who benefits. This is not yet a licensed assay-driven decision, but it strengthens the hypothesis that the chemotherapy benefit in these women is largely an ovarian suppression effect — and it argues that a young woman with biochemically depleted ovarian reserve is unlikely to gain from cytotoxics.

Turning to oncogene-driven lung cancer, where the theme is what happens after our best targeted agents stop working. Facchinetti and colleagues, also in Annals of Oncology, profiled 590 patients progressing on targeted therapy using paired tissue and liquid biopsies [6]. Among 312 patients progressing on first-line osimertinib, RAS alterations accounted for resistance in about eleven percent, with a striking enrichment for KRAS G12D and almost no G12C. In ALK-positive disease, RAS alterations were roughly three times more common after lorlatinib than after second-generation inhibitors, and across MET-, ROS1- and RET-driven cancers they appeared in seven to sixteen percent of cases. In patient-derived models carrying acquired KRAS G12D and G12R mutations, combining osimertinib with either a selective G12D inhibitor or a pan-RAS inhibitor was markedly synergistic in vitro and in vivo. Two practical messages: tissue and liquid biopsy were complementary rather than interchangeable, so if you are only sending plasma at progression you are missing mechanisms; and the RAS inhibitor pipeline now has a rational, testable indication beyond primary KRAS-mutant disease. Alongside that, Cancer Cell reports a phase one trial of SYS6010, an EGFR-targeting antibody-drug conjugate, in 236 patients with previously treated advanced non-small cell lung cancer [9]. Response rates ranged from about a fifth in EGFR wild-type squamous disease to roughly forty-six percent in EGFR-mutant disease treated with tyrosine kinase inhibitors but not yet chemotherapy. In the heavily pretreated EGFR-mutant group who had had both a tyrosine kinase inhibitor and platinum, median progression-free survival was 7.6 months and median overall survival 19.4 months. Toxicity was almost universal and myelosuppressive — grade three or higher events in just over half of patients, driven by neutropenia, leukopenia and thrombocytopenia — so this is a payload with a real haematologic cost, though notably not the interstitial lung disease signal we watch for with other lung antibody-drug conjugates.

Our third theme is chemotherapy-free treatment in lymphoma. In the Journal of Clinical Oncology, Lynch and colleagues report a response-adapted, entirely chemotherapy-free regimen for untreated indolent B-cell lymphoma [7]. Forty-two patients, mostly follicular lymphoma and mostly stage three or four, received eight cycles of the CD3-by-CD20 bispecific mosunetuzumab; those not in complete metabolic remission then went on to polatuzumab vedotin plus obinutuzumab. Every patient responded. Complete response was seventy-one percent after mosunetuzumab alone and eighty-six percent at end of treatment, with two-year progression-free survival of eighty-nine percent and no deaths. Cytokine release syndrome occurred in about two thirds OF PATIENTS but was uniformly grade one, and no one needed tocilizumab. The four progressions are biologically interesting — two from CD20 loss and two histologic transformations. This is a single-arm phase two study in forty-two patients, so it does not displace chemoimmunotherapy or the watch-and-wait conversation, but it is a credible template for randomised testing of a response-adapted, chemotherapy-free first line.

Finally, a cluster of papers about safety, trial interpretation, and guidelines. Clinical Cancer Research carries a commentary from Alumkal and Ellis on the withdrawal of the EZH2 inhibitor tazemetostat, after the confirmatory SYMPHONY-1 trial found haematologic second primary malignancies — mostly myelodysplastic syndrome and acute myeloid leukaemia — in about six percent of treated patients and none in the control arm [3]. Their argument is that this is not idiosyncratic but on-target: EZH2 is context-dependent, and loss-of-function states are recurrent drivers of myeloid neoplasia, so chronic pharmacologic inhibition may phenocopy them. That matters directly for ongoing phase three programmes in prostate cancer, where exposure is long and latency-dependent risk accumulates — a reason to be vigilant about cytopenias and clonal haematopoiesis in patients on any epigenetic agent. In the same journal, Shahnam and Gounder revisit BRIGHTLINE-1 in dedifferentiated liposarcoma, arguing that a trial reported as negative in fact identified two active drugs [8]. Brigimadlin produced a median progression-free survival of 8.4 months and responses in about a fifth of patients with translational evidence of p53 reactivation; the trial failed largely because doxorubicin substantially outperformed historical expectations, which resets the benchmark for future sarcoma trials. There is a shared lesson with the third paper here: a phase two trial in Clinical Cancer Research from Kelley and colleagues added granulocyte-macrophage colony-stimulating factor to pembrolizumab in 42 patients with advanced biliary cancer [10]. The confirmed response rate was twelve percent, including two complete responses, and about a quarter of patients were progression-free at six months — but the study did not meet its prespecified efficacy threshold. It was well tolerated, and paired biopsies showed increased CD8 T cells and antigen processing, so the immunologic hypothesis remains live even though this combination should not move forward as designed. And rounding out the week, the Journal of Clinical Oncology published two ASCO living guideline updates — one on stage four non-small cell lung cancer with driver alterations [1] and one on immunotherapy and targeted therapy for advanced gastroesophageal cancer [2] — both worth a look given how fast both fields are moving.

If you only have time for one paper this week, make it the RxPONDER ovarian reserve analysis in Annals of Oncology [5]. A single pretreatment hormone measurement outperformed menopausal status and age at identifying which young women with node-positive, low-recurrence-score breast cancer actually gain from chemotherapy — and that is a decision we make constantly.

Here are the key takeaways from this week in Oncology. Enzalutamide's survival advantage in metastatic hormone-sensitive prostate cancer persists at eight years, with fewer prostate cancer deaths, no excess non-cancer mortality, and most long-term patients holding the full dose. In premenopausal breast cancer with a recurrence score of 25 or less, low anti-Müllerian hormone identified women who derived no benefit from adjuvant chemotherapy, supporting the ovarian suppression hypothesis. RAS alterations drive resistance in roughly one in ten oncogene-driven lung cancers, they are enriched after the most potent inhibitors, and tissue plus liquid biopsy together find more than either alone. A chemotherapy-free, response-adapted bispecific-based regimen achieved high complete response rates in untreated follicular and marginal zone lymphoma with only grade one cytokine release syndrome. And on interpretation and safety: EZH2 inhibition carries a plausibly on-target myeloid malignancy risk that demands vigilance in long-duration trials, while both BRIGHTLINE-1 and the biliary immunotherapy trial remind us that a missed endpoint is not the same as an inactive drug — or an uninformative study.

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 a quick rating on Apple Podcasts or Spotify helps other physicians discover the show.

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