Chest wall recurrence outcome by initial stage
ForPost-mastectomy isolated CWR, no prior PMRT, no distant mets
TL;DR5-yr OS after isolated CWR 79.9% vs 41.9% vs 29.1% (T1-T2N0 vs T1-T2 1-3 N+ vs T3-T4/>4 N+); 1-3 N+ salvage as poor as advanced stage.
The RT read has two parts. Omitting RT for the CWR carried OS OR 4.524 on MVA, and pts with T1-T2 1-3 N+ disease salvaged no better than T3-T4/>4 N+ (5-yr OS 41.9% vs 29.1%). The PMRT decision in 1-3 N+ therefore rests on preventing CWR, not on rescuing it. Salvage dose and field are not reported.
In T1-T2 breast cancer with 1-3 positive nodes after mastectomy, this supports counting salvage of a chest wall recurrence as unreliable when weighing PMRT; it does not extend to T1-T2N0 pts, whose 5-yr OS after CWR was 79.9%.
Omitting RT for the CWR carried OS OR 4.524 on MVA, and pts with T1-T2 1-3 N+ disease salvaged no better than T3-T4/>4 N+ (5-yr OS 41.9% vs 29.1%). The PMRT discussion in 1-3 N+ should treat salvage as unreliable. Salvage dose and field are not reported.
17 details 1 trial watching
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).
| Feature | T1-T2N0 | T1-T2 1-3 N+ | T3-T4/>4 N+ | p |
|---|---|---|---|---|
| LVI present | 8 (19.5%) | 8 (40.0%) | 22 (62.9%) | <0.001 |
| Neoadjuvant therapy | 2 (4.9%) | 0 (0%) | 14 (40.0%) | <0.001 |
| Adjuvant systemic therapy | 15 (36.6%) | 17 (85.0%) | 27 (77.1%) | <0.001 |
| XRT for CWR | 29 (70.7%) | 15 (75.0%) | 24 (68.6%) | 0.879 |
| Surgery for CWR | 29 (70.7%) | 9 (45.0%) | 16 (45.7%) | 0.045 |
| Factor | OR (95% CI) | p |
|---|---|---|
| Group 1 vs 2 | 4.871 (1.756-13.510) | 0.002 |
| Group 1 vs 3 | 7.521 (2.761-20.482) | <0.001 |
| Group 2 vs 3 | 1.544 (0.645-3.699) | 0.330 |
| No XRT for CWR | 4.524 (2.064-9.918) | <0.001 |
| Interval to CWR ≤24 mo | 4.636 (2.106-10.206) | <0.001 |
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.
- Does PMRT improve OS in T1-T2 with 1-3 positive nodes? recruiting The T-REX Trial: Tailored Regional External Beam Radiotherapy in Clinically Node-negative Breast Cancer Patients With 1-2 Sentinel Node Macrometastases. Phase NAn=1350 · primary completion 2028-12 · RCT omitting regional RT, T1-2 cN0, 1-2 SLN macromets
- Does salvage outcome after CWR hold with modern systemic therapy?
- Optimal salvage RT dose and field for isolated CWR
📚 Sources · 📄 1 paper
The longer read
The argument this paper makes is indirect but useful. Whether PMRT improves survival for a subgroup depends on three quantities: the baseline locoregional risk, the toxicity of treatment, and how well a recurrence can be salvaged. The third is usually waved away with the assumption that an isolated chest wall recurrence can be treated when it appears. For node-negative disease this series supports that assumption, with 5-yr OS after CWR of 79.9%. For T1-T2 tumors with 1-3 positive nodes it does not: 41.9% at 5 years, statistically indistinguishable from the T3-T4 or >4 node group that has long been irradiated by default. If the comparison holds, a recurrence in the intermediate group is a marker of systemic failure as much as a local event, and the case for PMRT there is about prevention rather than rescue.
The methodology limits how far that can be taken. The key group has 20 pts, so the null contrast between Groups 2 and 3 (OR 1.544, 95% CI 0.645-3.699) is compatible with a meaningful difference in either direction. Calling the two groups equivalent overstates what the confidence interval allows. The cohort is also selected: it contains only pts who recurred, and the advanced-stage pts who skipped PMRT did so for reasons often recorded as refusal, physician bias, or unknown. The arithmetic needs checking too. Cox estimates are labeled odds ratios, the Group 3 5-yr OS reads 29.1% in the abstract but 21.8% in Table 2, and the univariate DMFS pairwise estimates repeat the OS figures to three decimals. None of these overturns the direction, but they argue against quoting the exact magnitudes.
The salvage RT finding needs the most caution. Omitting RT for the CWR carried an OR of 4.524 for death on multivariable analysis, but the reasons RT was withheld include progression on systemic therapy, so the variable partly encodes biology that was already bad. It is reasonable support for irradiating an isolated CWR, which was already standard practice at the institution. It is not a clean estimate of what RT adds. Dose, field design and technique are not reported, so there is no transferable parameter for an RT reader.
The question the authors left to a randomized trial, whether PMRT improves OS in 1-3 node-positive disease, is the one that matters. This series cannot answer it. What it contributes is the reason the question deserves an answer: if salvage is unreliable, even a modest absolute LRR reduction could translate into survival, which fits the authors' note that pts accept treatment for OS gains as small as 1-2%. The paper would be wrong if the poor outcome in Group 2 reflects the era's systemic therapy or chance in a 20-patient group rather than stage biology. Both are plausible, and a reader should treat the result as a prior that tilts the PMRT conversation for this group, not as evidence that settles it.