The prognostic value of p53 and c-erb B-2 immunostaining is overrated for patients with lymph node negative breast carcinoma - A multivariate analysis of prognostic factors in 613 patients with a follow-up of 14-30 years

BACKGROUND. Approximately 30% of breast carcinoma patients with negative ly mph nodes die of their disease. Biologic markers such p53 protein and c-erb B-2 have been related to tumor progression, but their prognostic value rem ains controversial. METHODS. Two large series of a total of 613 lymph node negative breast carc inoma patients from a single institution were analyzed with respect to tumo r size, histologic grade, and immunohistochemical staining for p53, c-erb B -2, estrogen receptor (ER), and progesterone receptor (PgR). Interobserver variation in histologic grading was evaluated by Kappa statistics. The two series had different treatment modalities: 228 patients (SACGS group) were treated surgically with mastectomy and given 1 perioperative chemotherapy c ourse, and 385 patients (HOST group) were treated with mastectomy and ovari an radiation and further randomized to receive postoperative treatment with radiotherapy or no adjuvant treatment. The follow-up ranged from 14-30 yea rs. RESULTS. Immunoreactivities for p53, c-erb B-2, ER, and PgR did not differ significantly in the two series. p53 immunostaining was present in 187 of 6 13 tumors (29%), and c-erb B-2 immunoreactivity was present in 58 of the tu mors (10%). Three hundred forty-eight tumors (57%) were positive for ER. Ka ppa statistics value of interobserver variation in the histologic grading o f ductal carcinomas was 0.69, which is considered to be a substantial degre e of agreement. No significant differences in survival were found when comp aring p53, c-erb B-2, ER, and PgR positive and negative cases. However, bot h recurrence free survival rates and overall survival rates after 10 years were significantly better in the T1N0M0 group compared with the T2N0M0 grou p (81% vs. 67% [P < 0.0001] and 85% vs. 70% [P < 0.0001]). Ten-year recurre nce free survival rates for patients with histologic Grade 1 versus Grades 2-3 (according to Elston and Ellis' modification of the Bloom and Richardso n method) tumors were 90% and 70%, respectively (P < 0.0001), and overall s urvival rates for the same groups were 94% and 81%, respectively (P=0.0002) . After 30 years of follow-up, the overall survival rate for patients with tumors of histologic Grade 1 versus Grades 2-3 were 87% and 68%, respective ly, and were 78% and 66%, respectively, for patients with tumors less than or equal to 2 mm versus those with tumors > 20-50 mm. Approximately 35% of the patients with turners of histologic Grades 2-3 and measuring > 20 mm we re dead after 10 years of follow-up, contrary to 6% of the patients with tu mors of histologic Grade 1 measuring less than or equal to 20 mm. A signifi cantly more favorable prognosis also was observed in patients in the HOST g roup treated with adjuvant radiotherapy. CONCLUSIONS. Histologic grade and tumor size were found to be major prognos tic factors for patients after 30 years of follow-up. c-erb B-2 and p.53 im munostaining.