ASTRO 2026 - Summary of clinically relevant trials

[Gfunk6]

Advertisement - Members don't see this ad

ASTRO 2026 clinical update

1 Resected brain metastases: Alliance A071801​

Fractionated SRS makes more sense for the larger postoperative cavity. Alliance A071801 randomized 242 patients after resection of a brain metastasis >2 cm. One-year surgical-bed control was 87% with fractionated SRS versus 81% with a single fraction, with no significant toxicity difference. The trial used 27 Gy in 3 fractions or 30 Gy in 5, depending on target volume. I’d favor fractionation for these patients as volume and OAR constraints allow. The unexpected survival advantage needs explanation; the local-control benefit is the reason to change. [1]

2 Small cell lung cancer with brain metastases: NRG-CC009​

We now have randomized data supporting SRS for selected patients with small cell lung cancer brain metastases. NRG-CC009 randomized 151 patients to SRS or HA-WBRT plus memantine. Cognitive failure, the primary endpoint, did not differ. Median OS was 17.4 versus 8.6 months favoring SRS, but OS was secondary. Lesions could be up to 4 cm, with total volume up to 30 cc; median lesion count was two. I’d offer SRS when close MRI surveillance and salvage are realistic. SCLC alone should no longer dictate whole-brain RT. [2]

3 Small cell lung cancer without brain metastases: MAVERICK​

MAVERICK enrolled 304 patients with SCLC and a negative brain MRI after initial treatment; 68% had limited-stage disease. MRI surveillance improved cognitive failure-free survival (HR 0.60), including compared with HA-PCI. I favor surveillance when we can deliver the trial’s follow-up: MRI every three months in year one, then every six months in year two, with prompt salvage. PCI still reduces brain metastases and remains reasonable when surveillance is impractical. Final OS noninferiority results are pending, so survival equivalence is not yet established. [3]

4 WHO grade 2 meningioma after complete resection: ROAM/EORTC-1308 *technically not ASTRO, but relevant*​

ROAM/EORTC-1308 gives us randomized evidence for adjuvant RT after complete resection of a WHO grade 2 atypical meningioma. In 157 patients, 60 Gy in 30 fractions improved five-year DFS to 79.9% versus 64.3% with observation (HR 0.51; p=0.0396). Recurrence occurred in 14% versus 30% at a median follow-up of 64 months. Grade 2–3 radiation-related serious adverse events occurred in 5 of 66 treated patients (8%). I’d offer adjuvant RT after gross total resection and discuss the roughly 16-point DFS benefit alongside the risk of late toxicity. This was reported at EANO 2026 and published in The Lancet. [4]

5 Localized prostate cancer: PACE-A​

PACE-A is useful when a man with localized prostate cancer is choosing between surgery and SBRT. The trial randomized 123 surgical candidates, nearly all with intermediate-risk disease, to prostatectomy or 36.25 Gy in 5 fractions without ADT. At eight years, freedom from biochemical/clinical failure was 91% with SBRT versus 84% with surgery, without a significant difference. Among five-year questionnaire respondents, pad use was 8% versus 48%. The urinary continence difference belongs in the consult. This small trial did not establish oncologic superiority or noninferiority. [5]

6 Prostate cancer after prostatectomy: NRG-GU003​

GU003 is about prostate-bed RT after prostatectomy for prostate cancer. It compared 62.5 Gy in 25 fractions with 66.6 Gy in 37. Disease control and five-year patient-reported outcomes were similar, but grade 3 GU events were 13% versus 3%, mainly cystitis and hematuria. Five-year EPIC completion was only 56%. A ten-point difference in severe urinary toxicity matters even when the questionnaires look similar. I’d continue to offer 25 fractions, include this result in consent, and keep conventional fractionation available, particularly for men with significant baseline urinary problems. [6]

7 p16-negative head and neck squamous cell carcinoma: NRG-HN009​

HN009’s p16-negative cohort randomized 234 patients receiving definitive RT to 70 Gy with cisplatin 100 mg/m² every three weeks or 40 mg/m² weekly. Weekly cisplatin did not reduce overall grade 3–4 toxicity burden, and this cohort failed the toxicity criterion for phase III continuation. Six-month locoregional failure was 9.6% with weekly versus 4.6% with every-three-week treatment. Weekly caused less AKI and dehydration but more leukopenia. I would not routinely switch a fit p16-negative patient to weekly treatment on the assumption that it is gentler. Efficacy noninferiority was not established. [7]

8 Breast cancer after mastectomy: CHN HYPOPMRT​

CHN HYPOPMRT supports three weeks of PMRT for appropriately selected patients with high-risk breast cancer. With median follow-up of 11.5 years in 810 analyzed patients, ten-year locoregional recurrence was 12.0% with 43.5 Gy in 15 fractions versus 10.3% with 50 Gy in 25, without significant survival differences. No brachial plexopathy was reported. Grade 1–2 pulmonary fibrosis was higher with hypofractionation, 19.7% versus 11.7%. The fields were chest wall and supraclavicular nodes after ALND. High-risk disease alone is not a reason to insist on five weeks. [8]

9 Pancreatic adenocarcinoma before surgery: SOFT Preop​

For pancreatic adenocarcinoma, SOFT Preop compared preoperative SBRT with 50.4 Gy in 28 fractions plus concurrent chemotherapy after induction chemotherapy. Among 102 randomized patients, node positivity was 41% versus 38%; SBRT better preserved physical function and reduced fatigue without more surgical complications. Median OS was 26 versus 43 months (p=0.639). That is not evidence of survival equivalence. Five fractions is reasonable when the MDT has already chosen preoperative RT and we can reproduce the coverage and GI protection. The trial used 33–40 Gy in 5 fractions to the primary and 25 Gy in 5 to elective nodes; most SBRT patients had MR adaptive guidance. It did not answer whether to add RT. [9]

10 Unresectable or recurrent hepatocellular carcinoma: NRG-GI003​

GI003 compared protons with photons for unresectable or recurrent hepatocellular carcinoma in 115 eligible patients. It crossed the futility boundary for proton OS superiority. Two-year OS was 56% with protons versus 69% with photons (HR 1.30), with no significant PFS or local-control advantage. Grade 3+ toxicity was 11% versus 24%, a nonsignificant difference (p=0.09). If a photon plan meets constraints, I see no reason for an automatic proton referral based on the diagnosis. Referral should depend on a meaningful advantage on comparative planning. These results do not prove proton inferiority. [10]

11 Newly diagnosed glioblastoma: TRIDENT​

TRIDENT tested starting TTFields during chemoradiation in newly diagnosed glioblastoma. The 981 patients were randomized to start the device with RT/TMZ or with maintenance TMZ; both arms planned maintenance TTFields. Early initiation did not improve OS: 17.7 versus 17.5 months (HR 0.95; p=0.52). Device-related events were more common, 76% versus 59%, mostly skin reactions. I would not start TTFields during RT on these data. The post hoc MGMT-methylated, compliant subgroup does not rescue the negative primary endpoint. This addresses when to start the device; it did not test maintenance TTFields against treatment without it. [11]

References​

1. Brown PD, Ballman KV, Aizer AA, et al. Alliance A071801: postoperative single-fraction versus fractionated SRS for resected brain metastasis. ASTRO 2026, LBA 01.

2. Rusthoven CG, Paulus R, Gondi V, et al. NRG-CC009: SRS versus hippocampal-avoidant WBRT plus memantine for SCLC brain metastases. ASTRO 2026, LBA 02.

3. Rusthoven CG, Redman M, Brown PD, et al. SWOG S1827/MAVERICK: brain MRI surveillance with and without PCI for SCLC. ASTRO 2026, LBA 03. See also ASTRO Daily News, Clinical Trials session highlights, September 28, 2026.

4. Jenkinson MD, Rosala-Hallas A, Sahm F, et al. Radiotherapy versus observation following surgical resection of WHO grade 2 atypical meningioma (ROAM/EORTC-1308): phase III randomized trial. Lancet. Published September 25, 2026. doi:10.1016/S0140-6736(26)01804-0. Presented at EANO 2026.

5. van As N, Patel J, Tree A, et al. Efficacy of radical prostatectomy versus SBRT for localized prostate cancer: PACE-A long-term outcomes. ASTRO 2026, LBA 32.

6. Buyyounouski MK, Borges M, Chen RC, et al. Five-year results of hypofractionated versus conventional postprostatectomy RT: NRG Oncology GU003. ASTRO 2026, Abstract 7.

7. Harari PM, Attwood K, Siddiqui F, et al. NRG-HN009: every-three-week versus weekly cisplatin with definitive RT, p16-negative phase II toxicity results. ASTRO 2026, LBA 06.

8. Sun GY, Chen SY, Yu T, et al. Hypofractionated versus conventional PMRT in high-risk breast cancer: ten-year outcomes from CHN HYPOPMRT. ASTRO 2026, Abstract 1.

9. Hall WA, Tsai S, Banerjee A, et al. SOFT Preop: randomized phase II comparison of preoperative SBRT and conventionally fractionated chemoradiation for pancreatic adenocarcinoma. ASTRO 2026, LBA 08.

10. Hong TS, Winter K, Koay EJ, et al. Initial results of NRG-GI003: phase III randomized comparison of protons and photons for hepatocellular carcinoma. ASTRO 2026, Abstract 2.

11. Shi W, Glas M, Lapointe S, et al. TRIDENT: concurrent versus maintenance initiation of TTFields with temozolomide-based therapy for newly diagnosed glioblastoma. ASTRO 2026, LBA 04.

 

[TheWallnerus]

6 Prostate cancer after prostatectomy: NRG-GU003​

GU003 is about prostate-bed RT after prostatectomy for prostate cancer. It compared 62.5 Gy in 25 fractions with 66.6 Gy in 37. Disease control and five-year patient-reported outcomes were similar, but grade 3 GU events were 13% versus 3%, mainly cystitis and hematuria. Five-year EPIC completion was only 56%. A ten-point difference in severe urinary toxicity matters even when the questionnaires look similar. I’d continue to offer 25 fractions, include this result in consent, and keep conventional fractionation available, particularly for men with significant baseline urinary problems. [6]

Me offering 1.8Gy vs 2.5Gy fractions for prostate cancer: