POST-MASTECTOMY RADIATION

Sites of failure after MRM:

1. Chest wall 60-80%
2. Supraclav
3. The failure rate in an adequately dissected axilla (≥10 nodes) is very low (~2%)

Risk of Local failure by node status

Risk Classification
Summary

Number of positive nodes Local Failure Rate in 10
0
----- < 5 cm tumor 9.2%
----- 5+ cm tumor 15% to 60%
1-3 5-31%
4+ > 20%
10+ 55%

Trials Data

Isolated Locoregional Failures (%)
Trial Systemic Treatment 1-3 Positive Nodes 4+ Positive nodes Follow-Up Years
Dana-Farber CMF 5 5
CAF ~ 20
82b CMF 30 42 10
82c T 31 46 10
SECSG CMF 20 10
ECOG CMF ect. 13 29 10
MDACC FAC 10 4-9+/21 10+/22 10
NSABP 90% doxorubicine based 13 4-9+/24 10+/32 10
British Columbia CMF 21 41 20

Clinical and pathologic factors predicting high (ie, > 20%), moderate (10% to 20%), and low-risk (< 10%) for LRF are used to categorize the potential locoregional benefit from comprehensive PMRT.

LOW RISK (<10%: pT1-pT2, pN0)

  • In old meta-analysis absolute improvement in local control with PMRT was outweighed by increase in vascular deaths.
  • EBCTCG meta-analysis 2000: PMRT reduces LRF from 9.2% to 2.7%, however it does not significantly reduced overall recurrence rate (30.2% to 28.2%).
  • NSABP B-04 25-yr update showed that PMRT had lowered local and regional recurrence without OS benefit.
Arms Local Recurrence Regional Recurrence
radical mastectomy 5% 4%
total mastectomy w/o ALND but w/ PMRT 1% 4%
total mastectomy with ALND. 7% 6%
  • Based on low absolute risk of locoregional failure and the small absolute improvement in locoregional control, contemporary trial do not include these low-risk women in PMRT trials.

MODERATE RISK (pN1a or pN1c)

Randomized trials British Columbia and Danish 82b and 82c

  • report 21-31% locoregional recurrence rates
  • Subset analysis of patients with 1-3 + nodes and ≥10 nodes dissected shows improved survival (by 18%) (an argument against Fowble and those who would not irradiate this group).
  • In the Danish trials 1-3 nodes+ benefit relatively more than 4+ nodes
  • Recurrences after mastectomy are not in general curable (so need to cure it at the time of initial treatment).

Evidence for lower locoregional failure (10-15%)

  • (15 year) Local failure rate in unirradiated patients with 1-3 nodes in 82b, 82c and British Columbia —30%,31%, & 33% — were substantially higher than in 5-yr FU reports – 6% to 13%. [Recht ECOG: JCO 17:1689-1700, 1999. Goldhirsch JCO 6: 89-97, 1988. Buzdar Cancer 65:394-399, 1990. Kaufmann JCO 11:454-460,1993.]
  • ECOG reported 2016 patients in 4 randomized trials using CMF + prednisone + TAM (> 1 year) + anthracycline based chemotherapy. The risk of LRF-10 was 12.9% in patients with 1 to 3 positive nodes. (Recht, JCO 1999;17:1689-700) This is also more consistent with US institutional data.

Possible reasons for high locoregional failure rates in the British Columbia and Danish trials

  • limited axillary dissection (7 nodes in Danish and 11 in British Columbia)
  • in Danish 82c 1 year to TAM vs 5 years that is typically used
  • in Danish 82b CMF was used vs anthracycline chemotherapy that is currently popular.

Other criteria that are being studied

  • # nodes removed
  • % nodes involved (> 25%)
  • Extracapsular extension
  • Lymphovascular invasion

HIGH-RISK

  • Three randomized trials shows about 10% OS-10 advantage with PMRT in all women with
    • LN+
    • > 5 cm tumor
    • skin or chest wall invasion
    • Locoregional advantage is bout 15-25%
  • In early intact breast cancer EBCTCG meta-analysis showed significant OS advantage with in patients with LN+ for trials started after 1975. (Ragaz Lancet 2000;356:1270)

Historical Reviews & Meta-analysis

  • 40+ randomized trials since 1940’s
  • Cuzick 1987/1994 – studies prior to 1975, most pts node -, orthovoltage. No change in OS, Øbreast cancer deaths with RT
  • Auqier 1992: Oslo II and Stockholm trials – modern MV and dose. No chemo. In node +, trend towards ≠OS (p=0.06), ≠Freedom from DM
  • Oxford (Early Br. Ca. Trialist Coll Grp): studies prior to 1985. Improved RFS but no improvement in OS (Ø Br Ca deaths)

RANDOMIZED TRIALS OF PMRT

TREATMENT FIELDS

Bibliography
1. Marie Overgaard, et al; N Engl J Med 1997;337:949-55. Postoperative radiotherapy in high risk premenopausal women with breast cancer who receive adjuvant chemotherapy.
2. Joseph Ragaz, et al. N Engl J Med 1997;337:956-62. Adjuvant radiotherapy and chemotherapy in node-positive premenopausal women with breast cancer.
3. Lancet 1999;353:1641-8
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-Share Alike 2.5 License.