Local recurrences following mastectomy: support for the concept of tumor dormancy

Local or regional recurrence of breast cancer occurs in 5%-30% of patients treated by Halsted radical or modified radical mastectomy. Lag time between treatment and recurrence varies widely, and it is not known whether the recurring tumor grows at a constant growth rate or at a more rapid rate after a period of tumor dormancy.

This study was undertaken to discriminate between the above-mentioned hypotheses, i.e., determine whether a tumor that recurs after mastectomy grows at a constant rate or whether it grows rapidly following a period of tumor dormancy.

A series of 122 patients with local recurrence as a first event after mastectomy for resectable breast cancer was evaluated. We measured the diameter of the recurring tumor (Dr) in each patient and calculated the diameter that the recurring tumor could have reached at the immediately preceding physical examination (Dpe), when no local relapse had yet been detected, by assuming an exponential growth during the treatment-free interval. For patients who had a calculated diameter Dpe that was large enough to have been detected at the previous examination, we assumed that a tumor 5 mm in diameter had been mistakenly missed, and the expected corresponding tumor diameter at the time of detection (Drc) was calculated. Finally, the minimum growth rate (mGR) consistent with the sequence "no detection-->recurrence of diameter Dr" was obtained by assuming an exponential growth from the tumor volume corresponding to a diameter 1 mm less than the diameter detection threshold.

A wide overlap between Dr and Dpe values was observed. Seventy-two (59%) of 122 Dpe values were larger than the minimum Dr; 18 (15%) were even larger than the median Dr value. The difference between expected and observed detection rates was highly significant (P < .0001). Furthermore, when treatment-free intervals were longer than 4 years, the difference between median Dr and median Dpe values failed to reach statistical significance. The Drc values were significantly lower than the related Dr values, while the mGR values were significantly higher than the corresponding growth rates (paired sample t test: P < .001).

This study provides evidence that the hypothesis of uninterrupted constant growth of locally recurring breast tumors should be rejected, as it implies a statistically significant departure from observed data. Our results suggest that a period of tumor dormancy followed by more rapid growth could provide an alternative and more reasonable description of tumor recurrence.