onc brain

Retrospective single-institution review of 789 mastectomy pts at Fox Chase; 136 (17%) had tumor within 1cm of the deep margin. 34 pts met criteria and received no postoperative RT. Median f/u 59 mo (range 7-143).

Invasive breast cancer, clinical T ≤5cm, 0-3 positive nodes, microscopic deep margin ≤1cm, MRM in all. Median age 43 (29-76); 65% node-negative, 35% 1-3 positive nodes; no deep muscle involvement.

Chemo ± tamoxifen in 23 pts (CMF 12, CAF 10, AC 1); tamoxifen alone in 7. Systemic therapy tracked age: 91% of women ≤50 got chemo vs 25% of women >50.

5 chest wall recurrences, median interval 26 mo. Cumulative incidence 9% at 5yr, 18% at 8yr. Age ≤50 vs >50: 28% vs 0% at 8yr (p=0.04); margin ≤2mm vs 2.1-6mm 24% vs 7% (p=0.36). 5yr OS 82%, DFS 74%.

The source cites Ahlborn (margin ≤4mm 5.7% vs >4mm 3%) and Mentzer (≤5mm 11% vs >5mm 3%), neither significant and both crude rates at short f/u. This series adds the age interaction those did not report.

Age band estimates rest on 11-12 pts per group. 19 of 34 margins were not centrally re-reviewed. Systemic therapy is pre-taxane, pre-HER2-targeted, so absolute risks likely overstate contemporary failure rates.

The title's claim is really two claims: margin alone does not drive risk, and young age does. The first rests on a non-significant comparison in 34 pts, which is absence of evidence rather than evidence of no effect.

Retrospective single-institution series, 5 events; age subgroup confounded with chemo use (91% vs 25%); no RT comparator arm, so PMRT benefit inferred not measured.

Single-institution retrospective review, Mallinckrodt Institute and affiliates, 1964-1986, N=224 irradiated with curative intent. Median follow-up 46 mo (range 24-241); actuarial life-table method.

First isolated local-regional recurrence after radical (34%), modified radical (61%) or total (5%) mastectomy. Excluded prior postmastectomy RT, pts failing salvage elsewhere, and clinical or radiographic distant mets. Chest wall alone in 134 (60%).

Entire chest wall via tangents, 1.5-6.0 MV to 4500-5000 cGy with bolus and a 6-12 MeV electron boost to gross disease; SCV AP field to 4500-5000 cGy at 3 cm. About 90% treated at 180-200 cGy/d; small 1-2 cm-margin fields used early in the series.

At recurrence, 49 pts received chemotherapy (most often CMF or CAF) and 69 hormonal therapy, usually tamoxifen. Combined-modality vs RT-alone analysis deferred to a later report.

5-yr LRC 57%; whole-CW fields gave 75% vs 36% 5-yr freedom from CW re-recurrence vs small fields (p=0.0001). For <3 cm lesions, 100% controlled at ≥6000 cGy vs 76% lower; larger lesions about 50% even at 7000 cGy.

Grade 2 sequelae 9%, grade 3 1%, none fatal. Arm edema 11% vs 1% with vs without elective axillary RT; pneumonitis higher with a parasternal field (8% vs 2% tangents alone).

Toonkel and co-workers, cited here, found a survival advantage with comprehensive nodal RT, while Deutsch reported the opposite, with that elective-nodal group carrying more extensive disease. Chen saw 27% failure in an uninvolved site after involved-field RT, consistent with this series' case for elective CW and SCV coverage.

Small-field group was only 14 pts and clustered in the early era, so the field-size effect also captures changes in staging workup and systemic therapy. The elective CW p-value is printed as both .32 and .33, and dose was set by physician preference.

The pattern of failure argues the whole chest wall carries subclinical disease, which favors comprehensive fields over excision alone or small portals. Large, diffuse or combined CW-plus-nodal recurrences stay poorly controlled at any dose, and the authors point to hyperthermia, altered fractionation or systemic combinations in prospective trials.

Single-institution retrospective spanning 22 years; field size, dose and elective coverage chosen by era and physician preference, so selection and era effects confound every comparison.

Retrospective single-institution cohort of 1141 consecutive pts treated 2001-2012. Nomogram from Fine and Gray competing-risks regression in 623 unirradiated pts; risk tiers by decision tree (CHAID); internal bootstrap validation only. Median follow-up 74.9 mo.

T1-2, 1-3 positive axillary nodes, mastectomy with negative margin plus ALND (median 17 nodes dissected). Excluded neoadjuvant therapy, synchronous metastases, other malignancy. All received adjuvant chemotherapy, 90.5% anthracycline- or taxane-based; median age 47.

46-50 Gy in 23-25 fx, CT-based planning. Fields: ipsilateral chest wall, infraclavicular and supraclavicular; IMN only for central tumors, axilla only with ECE.

LRR as first event (chest wall or regional nodes), DR and BCM, all as cumulative incidence with death as a competing risk.

Nomogram C index 0.73. Risk tiers by score: low ≤10 (n=412), intermediate (n=385), high >18 (n=344). PMRT associated with lower LRR, DR and BCM only in the high-risk tier.

EBCTCG showed PMRT benefit in 1-3 node pts, but its control 10-yr LRR of 21% is well above this cohort's 6.0% 5-yr unirradiated rate, which is the gap a risk model tries to exploit. SUPREMO (NCT00966888) is the prospective test of selective PMRT in this setting.

PMRT recipients skewed to adverse features, so within-tier comparisons carry residual confounding. Era of low trastuzumab use (26.2%) and no capture of nodal size or margin width; a prior 242-pt nomogram (Wadasadawala) used different factors.

The model mostly formalizes features clinicians already weigh (T stage, node count, ER, grade, LVI) into a score. Its value is the low-risk tier's very low baseline LRR, which supports omission, but the benefit estimates need external validation before they guide individual decisions.

Retrospective, clinician-allocated PMRT; risk cutoffs data-derived and tested on the same cohort without external validation. Low-risk null rests on 2 events.

Single-institution retrospective chart review of pts seen for CWR at MD Anderson 1988-1998. Kaplan-Meier, log-rank and Cox regression; time 0 = CWR diagnosis. Median f/u 41.4 mo from CWR.

N=96 post-mastectomy pts with pathologically confirmed CWR, no prior PMRT, no distant mets at CWR. Groups: T1-T2N0 (n=41), T1-T2 with 1-3 N+ (n=20), T3-T4 or >4 N+ (n=35). Group 3 had more LVI (62.9%) and neoadjuvant therapy (40.0%).

XRT was part of CWR salvage in 70.7% / 75.0% / 68.6% (p=0.879). Of the 29 not irradiated, reasons included refusal (6), progression on systemic therapy (6), no referral (3) and unknown (10). Dose, fractionation and target volume are not reported.

OS and DMFS from the date of CWR.

5-yr OS 79.9% vs 41.9% vs 29.1% (p=0.0001); 5-yr DMFS 75.2% vs 33.6% vs 25.9%. Group 1 beat Groups 2 and 3, and Group 2 vs 3 did not differ (MVA OR 1.544, p=0.330). No XRT for CWR independently predicted worse OS (OR 4.524) and DMFS (OR 4.024).

The PMRT trials the authors cite showed ~30% → ~10% 10-yr LRR and a 9% absolute OS gain in high-risk populations, which supports PMRT for T3-T4 or ≥4 N+. This series adds the salvage half of the argument for 1-3 N+, where the authors describe the LRR risk without RT as ~10-15%.

Cox estimates are labeled 'OR'. Group 3 5-yr OS differs between abstract (29.1%) and Table 2 (21.8%), and the DMFS pairwise ORs duplicate the OS values. The benefit of salvage XRT is confounded by indication: pts progressing on systemic therapy were not irradiated.

If salvage of a CWR in 1-3 N+ disease is as poor as in advanced stage, then the PMRT decision has to rest on prevention, which means the absolute LRR reduction becomes the deciding number. The series cannot say what that reduction is worth in OS. The authors defer that to the randomized Intergroup trial.

Single-institution retrospective, n=20 in the key 1-3 N+ group, non-uniform salvage, internal table/text discrepancies; PMRT benefit inferred from salvage outcomes, never tested.