Projections for relapse are complicated by a number of factors, the most profound of which is the very definition of relapse itself.  The problem is that “relapse or recurrence” might be taken to include distant recurrence, regional recurrence, local recurrence, and in some cases contralateral breast cancer. 

 

The Overview (1) states “Recurrence was defined as first appearance of breast cancer at any site (local, contralateral or distant) as in previous overviews”. 

 

The NSABP (2) has used the following definition for DFS (in some ways a reciprocal for relapse) “an 'event' is defined as the first documented evidence of local, regional, or distant recurrence, recurrence of tumor in the ipsilateral breast following lumpectomy, second primary breast cancer, or death without recurrence of cancer.”

 

Intergroup (3) has defined DFS as  “Defined as time from registration to earlier of relapse, death, or diagnosis of a new breast cancer with patients still disease-free and alive censored at date of last follow-up evaluation.”

 

The Overview's proportional risk reductions for recurrence are for the risk of recurrence which included contralateral breast cancer and so this is the definition used by Adjuvant!. This is however not entirely satisfying, because contralateral breast cancers would be expected to be biologically distinct from the primary, and the predictions of treatment effects on the contralateral breast cancer should be different. Many clinical trial groups have recognized this important issue and have changed the reporting of DFS to include only distant, regional, and local recurrence, and not to include contralateral breast cancer. Alternatively they may report relapse free survival (RFS) with this more restricted definition.  A useful analysis that could be done by the Overview would be to report the effects of different treatment modalities on distant, local/regional, and contralateral breast cancer separately. 

 

Where these differences in definitions would lead to the greatest confounding effect is in patients with low risk Stage 1 breast cancer. In these patients many of the events included in the broad definition of DFS will be contralateral breast cancers. Also in this group of patients details such as whether they had lumpectomy, unilateral mastectomy, or contralateral mastectomy would have large effects on total recurrence risk, but little effect on mortality. 

 

It must also be noted that local radiation therapy can have profound effects of local and regional event rates.  The Overview shows that the regional radiation has profound effects on local and regional relapse, but only small effects on survival.  

 

Estimates of outcome in Adjuvant! are most directly derived from mortality SEER data. The SEER data includes information about mortality, but it does not supply data about recurrence. Adjuvant! calculates estimates for recurrence, by adding 14% to the mortality risk (to account for the risk of contralateral breast cancer and local/regional events unlikely to result in breast cancer mortality), and inflates the annual hazard of breast cancer mortality a factor of 1.6 for ER positive cases and 1.1 for ER negative cases to account for the delay between relapse and death. Why the difference for the inflation between ER positive and ER negative cases.  The reason is because ER positive cases tend to recur later (4) and after recurrence to have a better survival than ER negative cases (5). Thus many more ER negative cases will have relapsed early and died within 10 years, than ER positive cases. Thus ER negative breast cancer 10 year mortality rates will more closely equal relapse (particularly distant disease [metastatic] relapse rates), that for ER patients where 10 year mortality rates will be lower that relapse rates (particularly distant disease [metastatic] relapse rates).

 

Because of these confounding effects and the more indirect estimates Adjuvant! recommends that its projections for mortality be used, as they are more directly and cleanly calculated. However, if relapse projections are used the limitations of these estimates should recognized.  

 

If one wants to make estimates of the improvement in relapse rates unconfounded by effects on contralateral breast cancer, you should subtract  (the approximate risk of contralateral breast cancer) from the estimates of relapse in the “10 Year Risk:” scroll box.  The number to subtract is shown below.  For example if the "10 Year Risk" scroll box shows a risk of 23%, you would subtract 6% and then type 17 into the "10 Year Risk " scroll box. 

 

 

Base Line Relapse Risk

Subtract Contralateral Risk

15 - 25 %

6 %

26 - 45 %

5 %

46 - 55 %

4 %

56 - 65 %

3 %

> 65 %

2 %

 

 

(1) Anonymous: Polychemotherapy for Early Breast Cancer:An Overview of the Randomised Trials. Early Breast Cancer Trialists' Collaborative Group. Lancet 352: 930-42, 1998.

 

(2) Fisher B.  Brown AM.  Dimitrov NV. et al.  Two months of doxorubicin-cyclophosphamide with and without interval reinduction therapy compared with 6 months of cyclophosphamide, methotrexate, and fluorouracil in positive-node breast cancer patients with tamoxifen - nonresponsive tumors: results from the National Surgical Adjuvant Breast and Bowel Project B-15. Journal of Clinical Oncology.  8(9):1483-96, 1990.

 

(3) Mansour EG, Gray R, Shatila AH, et al.: Survival advantage of adjuvant chemotherapy in high risk node-negative breast cancer: Ten year analysis- An Intergroup Study. J Clin Oncol 16: 3486-3492, 1998.

 

(4) Saphner T, Tormey DC, Gray R, Annual hazard rates of recurrence for breast cancer after primary therapy. J Clin Oncol 14: 2738-2746, 

 

(5) Chang J, Clark GM, Allred DC, Survival of patients with metastatic breast cancer.  Cancer 97: 545-553, 2003.