onc brain

Single-institution retrospective comparison of 693 single-isocenter VMAT SRS plans: HyperArc (23-pt pilot + 350 pts) vs 320 consecutive prior mVMAT pts as historical control. Clinical outcomes evaluated only in the HyperArc cohort.

Plan cohort mixed metastatic and benign/vascular targets; 59.9% single-target, 12.0% with 6 to 22 targets. Clinical cohort: 107 pts with 377 brain mets, median age 58, mostly NSCLC, breast and melanoma.

No margin on GTV. 18-20 Gy in 1 fx, or 30 Gy/5 fx when any target exceeded 2-3 cm or involved deep structures. 10 MV FFF, HD120 MLC, CBCT with 6DOF correction.

Per-target conformity, gradient and falloff indices, mean brain dose, film PSQA against TG-218, planning and delivery times. Clinical: lesion-level LC (no 25% diameter increase) and grade 3+ CTCAE v5 toxicity.

Conformity equivalent; gradient and falloff favor HyperArc; mean brain dose 4.6% vs 5.1%. 1-yr LC 90.1%, 3 grade 3 seizures, no grade 4/5.

Builds on the Clark 2012 class solution and the Thomas 2014 low-dose-spill penalty that matched Gamma Knife plans. Here the automated tool reaches that expert-solution level without nested tuning structures.

Software version differs between eras (Eclipse 15.5 vs 13.6). Planning time measured from structure-set timestamps as a surrogate. LC depends on 47 tumors at risk at 1 year.

The case for automation is standardization, not a dosimetric leap: an expert manual program already sits near the ceiling. What remains unsettled is rotational-error underdosing of off-isocenter targets with zero margin, and which plan-quality metric actually tracks clinical outcome.

Retrospective historical-control dosimetric comparison; clinical outcomes single-arm HyperArc only with 5-mo median imaging f/u and toxicity scored by chart review.

Prospective single-arm trial, N=40, enrolled Feb 2017 to Aug 2019, registered NCT02886572. Median follow-up 15.2 mo. Funded by Varian Medical Systems.

4 to 10 untreated brain mets, age ≥18, KPS ≥70, life expectancy >3 mo, DS-GPA ≥0.5, largest lesion ≤4 cm, >2 mm from optic apparatus; brainstem lesions allowed. Excluded SCLC, germ-cell, lymphoma, leukemia, myeloma, leptomeningeal disease. Median age 61, KPS 90, 6 mets; NSCLC 50%, breast 27.5%; 11 had prior SRS/WBRT.

Single-isocenter VMAT, 1 mm PTV margin. 20 Gy for PTV <2 cm, 18 Gy for 2.0 to <3.0 cm; 5 Gy x5 to all lesions if any PTV >3 cm, any brainstem lesion, or brain V12Gy >20 mL. 24 pts (125 lesions) got 1 fraction, 16 pts (79 lesions) got 5; median beam-on 8 min.

Primary: OS, judged non-inferior if ≥17 of 40 pts outlived DS-GPA-predicted survival. Secondary: local recurrence, distant brain recurrence, neurologic death, AEs, QOL, steroid use.

22 pts outlived prediction; mOS 8.1 mo, 1-yr OS 35.7%. 1-yr local failure 5% (10 of 204 lesions); distant brain recurrence 35.8% at 12 mo. Neurologic death 20% of deaths.

AEs in 73%, all G1-2, mostly headache and fatigue; no G3-5 treatment-related AEs. Radionecrosis in 5 lesions in 4 pts; 2 pts needed surgery. MMSE, Trail Making, HVLT and FACT-Br unchanged at 3 mo.

mOS of 8.1 mo sits inside the range of prior multi-met SRS series summarized in the paper's Table 4 (medians 5.8 to 14.1 mo), most of them retrospective. The randomized SRS vs WBRT trials for 4 to 20 mets (NCT03550391, NCT03075072, NCT01592968) are the comparators this cannot replace.

The DS-GPA benchmark uses the original formulation, measures from first brain-met diagnosis, and was applied here from protocol SRS in a cohort where 11 had prior brain RT. LC and DBF exclude the 8 pts who died before first MRI, and the paper's own Discussion gives a 6-mo distant failure figure (58%) that conflicts with its Results (20%).

The OS endpoint mostly shows the technique does not shorten life against a prognostic index; it cannot show SRS beats WBRT. The more usable signal is the per-lesion one: 5-fraction plans held 98% LC with low necrosis despite larger volumes, which supports hypofractionating high-volume multi-met plans.

Single-arm, N=40, single institution; OS benchmarked against a DS-GPA prediction, not a randomized WBRT or multi-isocenter comparator. Too few events for LC or necrosis modeling.

Retrospective two-center series (Rome, Negrar), 31 consecutive pts, 204 BM. Outcomes by Kaplan-Meier from SRS; dosimetric accuracy from offline CBCT fused to planning CT with residual shifts applied.

Age ≥18, 4-10 BM <3 cm. Median age 60; lung 14, melanoma 8, breast 5. Concurrent systemic therapy in 9 pts (IO 4, targeted 5).

Single-isocenter DCA via Brainlab Elements, 10 noncoplanar arcs at 5 couch positions. 20-22 Gy for <2 cm, 16-18 Gy for ≥2 cm or near critical structures; PTV = GTV + 1 mm (176 lesions) or 2 mm (28).

1y/2y LC 89%/77% per pt, 94%/86% per lesion; mOS 16.1 mo. Coverage loss (V95 <95%) in 28 targets, driven by distance ≥39 mm (P=.0001) and volume <0.40 cm3 (P=.015).

No G3-4 toxicity. Necrosis in 5 (16.1%) pts, 11% per lesion at 12 mo, 2 with G2 motor deficit; individual V12Gy >10 cm3 predicted it (P=.01).

Roper et al showed coverage held at 0.5° simulated rotation but failed in over one-third of plans at 2°; here measured residuals stayed within 0.5 mm / 0.5° tolerance, which is why 1 mm sufficed. Necrosis tracking V12Gy matches the Korytko gamma knife series the authors cite.

The paper reports pt counts inconsistently (4-6/7-10 vs 4-7/8-10; 10 vs 12 pts with coverage loss). LC is not stratified by margin or by distance from isocenter, so the clinical consequence of the dosimetric miss is untested.

The case for 1 mm is a trade-off: the margin rescues small distal targets while Table 4 shows each added millimetre inflates a 1 cm lesion's volume steeply, feeding V12Gy. That trade-off only holds if intrafraction drift is caught and corrected.

Retrospective 31-pt series, median f/u 11 mo, no comparator arm; margin conclusion rests on dosimetric re-computation, not lesion-level clinical failure by margin.

Retrospective chart review of 100 pts treated at Princess Margaret 1982-1993; 94 analyzed after 6 lost to follow-up. Median follow-up 2.9y (range 0.2-10.4) in the 45 survivors.

Histologically confirmed SCC of the nasal skin; nasal vestibule SCC excluded. T1 60, T2 11, T4 7, TX 16; only 1 pt N1 at presentation.

35 Gy/5 for lesions <2 cm, 45 Gy/10 for 2-5 cm, 50 Gy/20 for >5 cm or bone/cartilage invasion. Orthovoltage in 81%, energy chosen by lesion thickness (75-100 keV <2 mm, 125 keV 2-4 mm, 250 keV >4 mm); 1-cm field-edge margin; no elective nodal coverage.

Primary: local failure-free rate. Secondary: nodal failure-free rate, OS, cause-specific survival, late toxicity.

No RTOG G4 late toxicity and no cartilage or bone necrosis. Skin pallor, atrophy and telangiectasia common 1-5y post-RT, progressive but usually asymptomatic.

The Mazeron nose and vestibule series cited by the authors reported 11% local failure for SCC, in line with the 10% 2y local failure here. A cited vestibule series reported 5y local relapse-free 80%, consistent with the authors' point that vestibule disease does worse and was rightly excluded.

Treatment-era orthovoltage technique and no formal cosmetic assessment, so the optimal schedule cannot be derived. Dose was chosen by size and frailty, confounding any dose-response read.

Authors now prefer 50 Gy/20 or 60 Gy/30 for large infiltrative lesions, since schedules under 4 weeks usually leave a visible cosmetic defect after 1-2y. Many elderly pts still accept 1-2 week courses and the cosmetic trade-off.

Retrospective single-institution series without a surgical comparator; toxicity and cosmesis not formally scored; short survivor follow-up in an elderly cohort.

Retrospective single-institution series, 51 lesions in 48 pts referred 1999-2005. Median follow-up 45 mo (range 4-102); LC by Kaplan-Meier.

Facial BCC (82.4%) and SCC (17.6%), mean diameter 1.5 cm (0.5-4 cm), thickness ≤3 mm in all; thicker lesions went to electrons. Median age 78; 95% radical, 5% adjuvant after lateral-margin-positive excision.

Ir-192 HDR contact BT via custom wax nasal molds (CT-planned) or bolus applicators on flat skin, 1 cm margin, prescribed at 3 mm depth. Standard 48 Gy in 12 fx of 4 Gy, 3 fx/wk, over 4 wk; 3 Gy x 17-18 and 7 Gy x 5 schedules also used.

5 relapses (9.8%) at 6-32 mo, all on the nose and all with three-tube applicators. Nasal tip LC 73% vs 95% on pyramid/wing; all salvaged surgically.

Acute G4 dermatitis in 11 pts (21%), healed within 3 wk; no late complications including chondritis. Cosmesis good or very good in all cases.

Source cites Guix et al. contact HDR-BT with 5-yr LC 98% at 60-65 Gy/1.8 Gy fx, and Rio et al. LDR interstitial BT with 92.5% radical LC. This series sits below those on the nose, driven by the tip.

Only 9 SCC, so no histology-specific read. Cosmesis was physician-graded and the G4 dermatitis rate differs between Results (21%) and Discussion (12%).

Authors attribute nasal-tip failure to undersized three-tube molds and poor catheter bending at the tip's lower surface, and now add catheters. A deeper prescription point was not excluded as a cause.

Retrospective single-arm series, 51 lesions, one institution; no comparator vs Mohs, electrons or orthovoltage. Nasal-tip failures tied to an early three-tube applicator.

Prospective, multicenter, observational, single-arm cohort; N=792 of 800 planned (80% power to exclude a 10% 5-yr LR). Median f/u 5.1y; cumulative incidence with death as competing event.

Women ≥70y, ER ≥50%, HER2-negative, ≤1cm grade 1/2 or ≤2cm grade 1, free margins, after BCS with no RT and no ET. Median age 73.6y.

Primary: 5-yr LR, a histologically confirmed invasive ipsilateral event. Secondary: contralateral BC, OS, BC-specific survival.

5-yr LR 14.8% (95% CI 12.3-17.5), exceeding the prespecified 10% threshold. 2 of 81 deaths BC-related.

PRIME II and CALGB 9343 supported RT omission in older women who received ET; TOP-1 tests the population that skips both. The trial's own benchmark is a national registry LR of 1.3% with RT, not a randomized arm.

Only invasive ipsilateral events counted, so DCIS recurrence is excluded from the primary. Registry comparator differs in era and selection, and ER+ low-grade recurrences accrue well beyond 5y.

Dropping both adjuvant therapies failed the study's own safety bar, while salvage remained mostly breast-conserving. The authors keep RT omission as a shared-decision option, but the result does not show which omitted therapy drove the excess.

Single-arm observational cohort against a historical registry benchmark; no concurrent RT or ET arm, and 5.1y f/u is short for low-grade ER+ recurrence kinetics.

Randomized 2:1, chemo + SBRT vs chemo alone, 75 enrolled and 74 eligible across 14 centers. Phase not stated in source.

Potentially operable (n=48): borderline resectable, or operable with tumor >4 cm, extrapancreatic extension, or nodal involvement. Inoperable (n=26): locally advanced or medically inoperable.

Three months of mFOLFIRINOX or gemcitabine + nab-paclitaxel by clinician choice, then reassessment for surgery. Resected pts received 3 more months adjuvant; unresected pts 3 more months systemic chemo.

SBRT 40 Gy in 5 fractions with fiducial marker guidance and real-time RT quality assurance. 39 of 49 (80%) assigned to SBRT initiated it.

Primary: locoregional recurrence at 12 months. Secondary: PFS, OS, surgical outcomes, toxicity.

Primary endpoint met; benefit significant in the potentially operable stratum, not significant in inoperable disease (P=0.16). Resection and R0 rates numerically higher with SBRT, both non-significant.

G3+ toxicity 70.2% vs 64.0% (P=0.27), no significant increase with SBRT. RT-specific and late toxicity not broken out in source.

Randomized evidence for adding RT to chemo in pancreatic cancer has been inconsistent: LAP07 tested conventional chemoRT in locally advanced disease, and PREOPANC tested preoperative chemoRT in resectable and borderline disease. MASTERPLAN differs by pairing short-course SBRT with modern multiagent chemo and by powering on local control rather than OS.

Incomplete SBRT uptake (80%) and clinician-chosen chemo add noise to a small trial. The OS gain sits on a secondary endpoint with no HR reported in the source, and late GI toxicity is not described.

A local control benefit of this size, carried into PFS and OS, makes a larger confirmatory trial a priority. It does not yet settle which stratum benefits, whether the OS signal is real, or how SBRT should be sequenced with surgery.

Randomized with primary endpoint met, but N=74 across heterogeneous strata, 80% SBRT uptake, and OS/PFS secondary; phase not stated, conference trade-press source only.

Prospective single-institution series, 136 consecutive pts treated 1992-1997 in Barcelona. Minimum follow-up 12 mo; median follow-up not reported.

Pathologically confirmed facial BCC (102) or SCC (34), 73 primary and 63 recurrent after prior local therapy. Gross tumor left in place in most; margins positive in 128. No nodal or distant disease on CT head/neck and chest X-ray.

HDR Ir-192 via 117 custom PMMA molds (3-7 parallel catheters, 5 mm bolus layer) or 19 Brock applicators. 6000-6500 cGy in 33-36 fx of 180 cGy at 5 mm depth; lesions ≥4 cm boosted 1800 cGy/10 fx after a 3-wk rest to 7500-8000 cGy. Eye shielded with a 5 mm lead sheet in the mold.

All pts complete responders; 3 local recurrences (1/73 primary, 2/63 recurrent). 5y actuarial LC 98% overall, 99% primary, 87% recurrent; no regional or distant failure.

Erythema G1 in 117 and G2 in 19; 14 (10%) had G1 ulceration at 6 wk, all healed by 4 mo. Good/excellent cosmesis 98% at 6 mo; no severe late complications.

Primary-tumor LC of 99% sits alongside the 95% superficial X-ray result in the Lovett EBRT series and above Lovett's 77% with electrons. Daly's Ir-192 interstitial implants reached 96% LC but with 18% severe late effects, the toxicity gap the mold approach is designed to close.

Dose per fraction (180 cGy) far from modern short-course skin brachy, so the late-effect profile may not transfer. Recurrent-tumor LC timepoint reported inconsistently (5y vs 4y), and cosmesis was graded by the treating team.

The contribution is technical: a reproducible, afterloaded surface applicator that conforms to non-flat facial anatomy with a sharp edge gradient. It shows high control with conventional fractionation but does not settle how it compares head-to-head with superficial X-ray, electrons, or Mohs.

Single-arm single-institution series without a comparator; follow-up minimum only 12 mo with median unreported; comparisons rest on historic EBRT and implant series.

Retrospective single-institution series of 13 pts treated at the Christie Hospital up to July 1996. Technique description with three illustrative cases.

9 BCC, 3 SCC, 1 intraepidermal carcinoma. Lesion confined to pinna skin; dermal ultrasound confirmed cartilage not invaded in all but one.

HDR microselectron with single stepping iridium source via custom vacuum-formed mould. 45 Gy in 8 fx over 5 days prescribed to the surface, mould height set so tumour depth lies within the 80% isodose. Sandwich mould case reduced to 42.5 Gy.

Complete tumour resolution in all pts, barely perceptible scar. No recurrences and no late complications over minimum 18 mo f/u.

Moderate cutaneous erythema limited to the high-dose area, settled by the 6 wk visit. No discernible reaction on the contralateral pinna skin.

Authors position moulds against electrons (poor fall-off for thin lesions, cartilage necrosis risk), 30 kVp X-rays (dose inhomogeneity on curved surfaces, limited availability) and surgery (amputation or excision with graft). No head-to-head data.

Outcomes reported as a blanket statement without per-pt f/u, actuarial LC, or a validated cosmesis scale. Late cartilage toxicity can emerge beyond 18 mo, and the LQ-based BED rationale assumes complete repair between fractions.

The contribution is technique, not efficacy: ultrasound depth measurement plus tunable source height makes the plan conformal to sub-millimetre tumour thickness. Whether a higher late-tissue BED is tolerated because of the small volume is asserted, not shown.

Single-institution case series of 13 selected pts, no comparator, 18 mo minimum f/u too short for late cartilage effects; outcomes descriptive only.

Single-institution retrospective review, January 2007 to December 2018. 50 pts, 53 lesions, minimum 6 mo follow-up, median follow-up 15 mo (range 6-27). LC by Kaplan-Meier.

Histologically confirmed BCC or SCC, AJCC 8th T1, all NCCN high-risk by area H or area M location. Median age 80; selected by a multidisciplinary dermatology clinic, mostly for inability to tolerate or refusal of surgery.

40 Gy in 8 fractions twice weekly (BED 60 Gy, α/β 10) via Ir-192 HDR catheter flap, CT-planned in BrachyVision. CTV = visible tumor + 5 mm, PTV = CTV with no margin; 5 mm spacers protected the nasal mucosa, globe and gingiva.

1-yr LC 92%, 4 local recurrences, 3 in nose/eyelid/lip. 49 lesions (92%) had good/excellent cosmesis with complete remission.

Acute G1 60%, G2 35%, G3 5%, all after fraction 5. Chronic G2 telangiectasia in 2 pts (4%) at 12 mo, no chronic G≥3, no toxicity-related hospitalization.

Local control falls inside the 87%-100% range of prior retrospective HDR surface-mold series. The only randomized surgery-vs-RT trial in facial BCC cited by the authors found 4-yr failure 7.5% vs 0.7% favoring surgery, though RT there was mostly interstitial brachy, not surface HDR.

Median V100 85% falls short of the ≥95% PTV coverage goal, and depth was assessed mostly by palpation. Abstract and body disagree on acute toxicity grade, and the Discussion understates chronic G2 events.

The flap's value is conformity on irregular surfaces where surface molds and electrons struggle, yet recurrences clustered in exactly those sites. BCC and SCC got identical dose and margins, consistent with NCCN, the 2020 ASTRO guideline and the 2019 ABS consensus, so histology-specific dosing is still untested.

Single-arm retrospective series, surgery-unfit/refusing pts, 15-mo median f/u; no comparator vs Mohs or EBRT. Consistent with prior HDR surface-mold series.

Retrospective review of all Ir-192 HDR surface mould cases at one BC Cancer center, 2010-2017, 65 lesions in 59 pts. Median follow-up 8 months.

Median age 82 (45-97). BCC 54%, SCC 31%, plus melanoma, lentigo maligna, Merkel cell, cutaneous NHL and one keloid. Mostly head and neck (forehead/temple, nose, scalp, cheek/lip); 7 lesions palliative-intent.

Individualized Aquaplast + superflab mould, catheters ~1 cm apart, CT-based planning in BrachyVision. Most common 40 Gy/10 fx (48.2%); range 15 Gy/3 to 60 Gy/30 fx. Prescribed at 0.5 cm depth, CTV from clinical wire with no PTV expansion.

2-yr LC 84.9%, 2-yr PFS 71.5%, 2-yr OS 77.6%; CR 96.8% at 2-4 mo. Of 20 deaths, only two were from the skin malignancy. No dosimetric or histologic factor reached significance on univariate analysis.

No G3-5 acute or late toxicity. Acute G2 dermatitis 32% and ulceration 8%; one late G2 necrosis (2%) appearing around 10 months.

The authors place their LC within the brachytherapy literature range of 83.3% to 100% and against EBRT LC of 87-100% and Mohs at 95-99%. Gauden et al. reported LC 98% with 36 Gy/12 fx, but at similarly short follow-up and largely in a post-resection margin-positive setting. GEC-ESTRO grades the technique B.

Histology and intent are pooled (keloid, lymphoma, melanoma, palliative lesions). Toxicity was graded retrospectively from records and photographs, and cosmesis was not scored. Internal inconsistencies (median age 82 vs 83, follow-up 8 vs 9 mo, sex counts summing to 65 of 59 pts) lower confidence in the tabulated data.

The series shows the technique is deliverable and tolerable in a very old cohort, which is the realistic use case. It does not settle the durability of control, the right prescription depth or margin, or whether coverage metrics predict recurrence.

Retrospective single-institution series, n=59, mixed histology and intent; 2-yr estimates drawn from a median 8 mo follow-up with no comparator arm.

Phase II protocol, single cohort, N=109 assessable of 111 treated, 304 irradiated sites. Response assessed 4-6 wk post-RT then every 3 mo; median follow-up for pts in remission 7 mo.

Symptomatic recurrent B-cell indolent lymphoma; proven transformation and concurrent steroids or chemo excluded. Median age 62, median 2 prior regimens (range 0-11), bulky disease in 52% of pts with size data.

4 Gy as 2×2 Gy 48h apart (n=80) or 1×4 Gy (n=29), megavoltage photons or electrons. Target was the entire nodal area or the node with a 1.5 cm margin, involved sites only; 6 pts had total nodal irradiation.

in-field lymphoma control, response defined by local activity only, worst site counted. Secondary: TP (any progression) and TLP (in-field), Kaplan-Meier from first RT day.

ORR 92%, CR 61%, PR 31%. Median TP 14 mo, TLP 25 mo; CR pts TLP 42 mo. Re-treatment in 41 selected responders: RR 98%, CR 71%.

Toxicity virtually absent: one transient alopecia, one transient hyperpigmentation. Antiemetics for upper abdominal fields; allopurinol for bulky disease, no gout or tumor lysis.

Concordant with Girinsky et al (135 sites, 48 pts) on RR, CR and inferior bulky-site results, but 2yr local PFS in PR pts 46% there vs 10% here. FORT later tested this low-dose approach against conventional-dose RT in a randomized design.

Comparison with pts' own prior regimens is historical and confounded by line of therapy, with power falling as numbers drop. Prognostic CR signals (≤2 sites P=.01, ≤5 cm P=.006) sit amid many tests; mixed target-volume definitions left to physician discretion.

A local, repeatable, near-toxicity-free option for systemic disease, with response rates matching first-line systemic therapy in this cohort. It does not settle whether 4 Gy matches conventional dose for durability, particularly in PR or bulky sites.

Single-cohort phase II, no randomized comparator vs conventional-dose RT or systemic therapy; within-patient historical comparison, 7-mo median follow-up in remitters.

Phase 2, nonblinded, randomized noninferiority trial at one institution, 1:1, N=160 (81 SBRT, 79 MFRT). Stratified by tumor size, bone site and number of sites irradiated. Enrolled September 2014 to June 2018.

Pathologic cancer diagnosis, painful bone mets (score ≥2 of 10), age ≥18, life expectancy >3 mo; up to 3 fields allowed. Excluded prior RT to the site, untreated cord compression, pathologic fracture at the site, and recent radioisotope.

SBRT 12 or 16 Gy x1 by lesion size (4 cm cutoff) vs 30 Gy in 10 fx. PTV = GTV + 5 mm, no CTV; 2D/3D, IMRT or VMAT delivery. Prophylactic 7-day low-dose dexamethasone recommended for SBRT only.

Primary: pain response by international consensus criteria (pain score plus MED), noninferiority margin 10%. Secondary: MDASI QoL, CTCAE toxicity, local control, OS.

No toxicity differences between arms. G3 fatigue 9.9% vs 5.1% (P=.37), fracture 1.2% vs 0% (P=.99); MDASI scores improved in both arms without a difference.

Prior randomized trials found no pain difference between single and multifraction conventional courses, with more retreatment after 8 Gy. RTOG 9714 and the Dutch Bone Metastasis Study are the conventional-dose trials behind that equivalence. This trial escalates the single fraction but never tests it against 8 Gy.

Dexamethasone given asymmetrically to SBRT pts confounds early pain response. The 16 vs 12 Gy contrast is non-randomized and tracks lesion size. Five SBRT pts did not receive treatment due to insurance denial, and median OS of 6.7 mo leaves few pts at risk for late local failure.

The ITT data support noninferiority and a local-control advantage for SBRT; the superiority read depends on per-protocol denominators. The live question is 16 Gy vs 8 Gy in pts selected for longer survival, which needs a phase 3.

Single-institution randomized phase 2 with one-sided alpha 0.20; superiority signals are per-protocol amid heavy attrition, comparator 30 Gy/10 rather than 8 Gy. Authors call for phase 3.

Single-centre series of 94 pts treated with Axxent eBT, Sept 2015 to Oct 2017. Each eBT plan was replanned a posteriori for Ir-192 and Co-60 (282 plans); toxicity scored by RTOG. Mean follow-up 14 mo (4 to 25).

Postoperative endometrial cancer, mean age 65.9 (33 to 84), FIGO IA through IVA. 33 pts (35%) received chemo before RT.

eBT alone 25 Gy/5 fx (n=37, stage IA plus seven IB1) or pelvic EBRT 23 × 2 Gy then eBT 15 Gy/3 fx (n=57), twice weekly. Dose prescribed at 5 mm from a 2.5 to 3.5 cm cylinder; 50 kVp source, TG-43 planning, no RBE correction.

Only one G2 acute event. At 1 month, vaginal G1 5.3%, rectal G1 1.1%, urinary G1 2.1%, no G2.

PORTEC-2 established vaginal brachytherapy as adjuvant RT for high-intermediate risk disease using conventional HDR; this series does not test eBT against that clinical benchmark. A cited multicentre eBT series reported bladder V50% 11.5 ± 9.7% and rectum 17.4 ± 10.9%, in the same range as here.

Higher V150/V200 places hotter dose at the vaginal mucosa, where late stenosis would show and 14 mo cannot capture it. TG-43 without heterogeneity correction may misstate low-energy dose in tissue.

The OAR advantage comes from rapid dose fall-off at low energy, the same property that concentrates dose near the applicator surface. Whether uncorrected RBE translates into more late mucosal toxicity, or whether RBE-based de-escalation would cost control, remains open.

Single-centre uncontrolled series, 14 mo mean f/u; HDR comparison is replanning, not a treated arm; no late toxicity or local control data at meaningful follow-up.

Retrospective multicenter imaging review, 4 academic centers, staging scans read by a single abdominal radiologist. Each node transferred to a CT template by its relation to the aorta and renal arteries, with craniocaudal scaling for height.

90 pts, CS IIA/B or recurrent CS I seminoma after orchiectomy, nodes ≤5cm. 40 (44%) relapsed on CS I surveillance at a median 8.5 mo (range 3.2-50.4); 50 (56%) node-positive within 3 mo of diagnosis. 48 (53%) left-sided, 41 (46%) right-sided primaries.

Proposed CTV: 2.5cm lateral/posterior and 2.1cm anterior expansion of arterial vasculature, superior border 1cm above the right or 1.5cm above the left renal artery. Classen acetabular inferior border retained; NCCN 2cm GTV boost and setup margins added on top.

145 nodes, average 1.6 per pt, median size 18 mm. 99% of nodes fell inside the proposed expansion; nodal region was unrelated to CS at presentation (P = 0.16) or primary laterality (P = 0.15).

Classen extended fields (prospective phase II, 2003) are the NCCN- and European consensus-endorsed standard; this map supports Classen's acetabular inferior border but argues the T10/T11 superior border overtreats. MRC TE10 linked smaller fields to less acute GI toxicity and leukopenia. Martin et al. used a 1.4cm aorta/IVC expansion from non-enlarged nodal anatomy.

Single-timepoint CT in many pts, so original nodal position before reaching size criteria was estimated, not observed. Two atypical nodes (1.4%) were excluded from the expansion analysis. Relapses that would emerge outside the reduced field after treatment are unmeasured.

The case for smaller fields rests on second-malignancy risk in a young cohort, with extended fields linked to up to 50% more SMNs than PA strip fields in cited data. The trade the authors accept is a small out-of-field relapse risk, judged salvageable, against lower SMN risk.

Retrospective imaging-defined node map; proposed CTV never tested for relapse or toxicity. Authors themselves limit it to research use pending validation.

Prospective multicentre observational study, 30 German centres, enrolment April 1991 to March 1994. 94 enrolled, 87 assessable, median follow-up 70 mo. Analysed ITT, including 22 pts with major protocol violations.

Histologically confirmed pure testicular seminoma, Royal Marsden stage IIA (<2 cm) or IIB (2-5 cm) after high inguinal orchiectomy. Elevated AFP excluded; βHCG up to 200 IU/L allowed after year 1. 66 IIA, 21 IIB; median age 32.5 and 32.

AP/PA hockey-stick to para-aortic/paracaval and ipsilateral high iliac nodes, gross disease plus 2 cm margin, T11 to acetabular rim. 2.0 Gy/fx: 30 Gy for IIA, 36 Gy for IIB, no boost; 5 pts treated bilateral iliac.

Primary: relapse-free survival at 5 years. Secondary: disease-specific survival (expected at least 95% at 5 yr) and acute/late GI and skin toxicity.

Acute toxicity dominated by grade 1 nausea (53% IIA, 52% IIB); no grade 4. No late GI toxicity, no peptic ulceration, no second malignancy to date; one pt with G1 skin hyperpigmentation.

Prior stage II series used larger portals covering the entire ipsilateral iliac/inguinal chain and contralateral pelvis; this cohort shows the reduced field loses no pelvic control. The source places modified EP (9% relapse, 33% G3/4 leukopenia) and carboplatin alone (15% failure) as no better on control with more toxicity.

Study was not designed to test reduced vs extended portals, so the toxicity advantage is inferred against historic series. Seven pts lost to follow-up were excluded rather than censored, and the IIB cohort is too small for its own protocol-violation analysis.

Failures were all extra-field (mediastinal or lung) except one lateral para-aortic field-edge relapse, and all were salvaged, so DSS was 100%. The authors suggest room for dose reduction below 30 Gy in IIA, though no dose-response relationship was established.

Single-arm observational cohort with no randomised comparator vs chemo; IIB subset small (n=21). Single-arm gate caps the verdict despite 70-mo follow-up.

Multi-institutional phase I/II, single-arm, N = 62. Median follow-up 39.4 mo.

Intraprostatic radiorecurrence (IPR) after any form of RT for localized prostate cancer. GC results available in 31 pts.

Dose-escalated focal salvage HDR, up to 30 Gy in 1-2 fractions. 24 pts received one fraction and 38 received two.

Primary: acute RT-related G3+ GU and GI toxicity (CTCAE v4.03). Secondary: all-grade toxicity, QoL (EPIC-26, IPSS), bPFS, rPFS, LRFS, MFS and survival.

No acute G3 toxicity. Late G3 toxicity in 2 pts (3.2%); QoL changes did not meet MCID criteria.

Prospective salvage brachytherapy data after EBRT are sparse, with RTOG 0526 testing whole-gland salvage LDR in this setting. F-SHARP adds a focal, dose-escalated HDR approach and a genomic risk layer that prior salvage series lacked.

The GC analysis is restricted to 31 of 62 pts and the >0.85 cutpoint rests on n = 4, so it is a descriptive tail, not a validated threshold. The 1 vs 2 fraction comparison was not randomized and is underpowered for equivalence.

Toxicity is the settled part: focal dose escalation to 30 Gy in the re-irradiated gland produced no acute G3 events. Durability is the open part, since bPFS at 3 yr is 62.7% and the efficacy endpoints carry no comparator.

Single-arm phase I/II, N=62, toxicity primary objective; efficacy endpoints secondary and uncontrolled. GC signal from half the cohort.

Retrospective single-centre audit of 97 pre/post-treatment CBCT pairs from 20 pts. Margins were calculated from IFPM alone and with the updated BIR geometric uncertainty guidelines.

36.25 Gy in 5 fractions SBRT without intra-fraction tracking, with fiducial-based daily online CBCT IGRT. Current margins were 6 mm isotropic, 3 mm posterior. Mean pre- to post-imaging interval was 9 min.

magnitude and direction of IFPM and coverage by current margins. Secondary: margins derived from IFPM alone and from the BIR guidelines.

Averaged 3D RMS IFPM was 2.5 mm (4.2 mm), mostly posterior (70%) and inferior (63%). Current margins covered 92% of fractions, and 8 of 97 exceeded them, six posteriorly.

The authors report RL motion above comparable CBCT series, with AP and SI comparable. A larger cited margin study (n=205) found smaller systematic errors and similar random errors.

Delineation error, the dominant component, came from an MRI-only study and is probably smaller than real CT or fused MR/CT error. Deformation was set to 0, match error was assumed rather than measured, and AP and SI were not split by direction.

The guideline-derived numbers suggest lateral margins could shrink while the inferior margin grows, but the inferior increase rests heavily on one outlier contour. The authors advise checking outcomes in current trials before changing practice, given favourable control and toxicity with existing margins.

Single-centre retrospective geometric audit, n=20, no clinical outcome endpoint; proposed margin changes are untested against toxicity or biochemical control.

Single-institution consecutive series from a prospectively maintained registry, 854 pts treated 1998-2012. Median follow-up 11.3 yr (max 19). Not randomized; no comparator arm.

Biopsy-proven, clinically localized prostate adenocarcinoma across all NCCN risk groups: 266 LR, 238 FIR, 106 UIR, 244 HR. Median age 68, median PSA 8.6 ng/mL. ADT in 61%, mostly 1-6 mo.

70 Gy/28 fx at 2.5 Gy/fx, IMRT with daily IGRT (BAT ultrasound 94.8%). CTV prostate alone for LR, plus proximal SV for IR/HR; PTV margin 4-10 mm. Elective pelvic nodes in only 5 of 244 HR pts.

Whole-cohort 10-yr bRFS 71%, cRFS 86%, OS 69%, PCSM 6.8%. Of 125 clinical failures, 117 were distant and 8 local-only.

10-yr late G≥3 GU 2% (mostly urethral stricture) and G≥3 GI 1% (mostly rectal bleeding). One grade 5 event each in GU and GI. 119 pts (14%) developed a new cancer, lung most common.

Authors place this alongside RTOG 0415, CHHiP, PROFIT, HYPRO and the MDACC series, none with 10-yr outcomes at publication. RTOG 0415 used the same 70 Gy/28 fx with G3 toxicity 3% to 4% at 5 yr, under looser dose-volume constraints and without mandatory IMRT.

Toxicity was physician-reported from retrospective chart review, likely underreporting events. Institutional constraints (rectal/bladder V70 5%, up to 10%) are far tighter than NRG-GU005 allows, so the toxicity rates may not hold under trial or community constraints.

The schedule controls the prostate well over 10+ yr; the risk-group gradient is driven by distant failure. What is left open is whether HR outcomes improve with longer ADT and nodal coverage at this fractionation.

Single-institution nonrandomized cohort without a comparator arm; long follow-up adds durability to randomized hypofractionation data but cannot establish equivalence on its own.