Prognostic potential of flow cytometric S-phase and ploidy prospectively determined in primary breast carcinomas

In a prospective study of a consecutive breast cancer series accumulated in the period 1978–82, the S-phase fraction (SPF) and ploidy status were determined by flow cytometry performed on cell nuclei derived from samples of 580 primary tumors. Sixty percent of the tumors were non-diploid. After correction for debris the median SPF values were 7.3% overall, 12% for non-diploid tumors, and 2.9% for diploid tumors (2.6% when nodal subsets N2 and N3 and cases with metastases at presentation were excluded). The SPF values correlated both to tumor size (p=0.008) and to the number of positive axillary lymph nodes (p=0.03).At clinical follow-up in 1986, 467 unilateral breast cancer patients who had undergone radical treatment for cure could be evaluated with respect to the prognostic value of both the SPF value and ploidy status. The median duration of follow-up was then 59 months (range 2–90), and the median time-to-recurrence 24 months (range 2–69, n=137).At follow-up in 1991, 201/467 of the patients had died, the median duration of follow-up being 50 months (range 2–126) for the deceased, and 119 (range 6–148) for the survivors. In multivariate analysis (Cox's proportional hazards models), the strongest independent predictors of distant recurrence-free survival (DRFS) were the number of positive axillary lymph nodes (p<0.0001), the debris-corrected SPF value alone (p=0.003,versus p=0.05 for uncorrected value), and ploidy status combined with the corrected SPF value (p=0.0002). When age was taken into account, both the corrected SPF value and the ploidy-SPF combination were predictors of crude survival (p=0.006 and p=0.002, respectively).In univariate life-table analysis, the 5-year DRFS rate was 93% in node-negative (N0) cases with an SPF<7.3%, as compared to 80% in those with an SPF7.3% (p=0.005). Among node-positive cases, the prognostic value of the SPF was confined to those with 1–3 positive nodes, the 5-year DRFS rate being 68% in cases with an SPF<7.3%, as compared to 40% in cases with an SPF7.3% (p=0.01).Ploidy status and SPF were combined to form four groups: diploid & SPF<2.6% (DL), diploid & SPF2.6% (DH), non-diploid & SPF<12% (NDL), and non-diploid & SPF12% (NDH). Among node-negative patients, the DRFS rate fell from 95% in the DL group to 87% in the NDL group, with the DH group at an intermediate level, as compared with 74% (p=0.03) for the NDH group which accounted for the bulk of the early distant recurrences. Among patients with 1–3 positive lymph nodes, the 5-year DRFS rate was 68% in both the groups with low SPF values (DL and NDL), as compared with 45% in the DH group (p=0.03), and 37% in the NDH group (p=0.006).In this study, the flow cytometry SPF value, alone or in combination with ploidy status, yielded the most profound additional prognostic information, enabling both node-negative patients with a high probability of cure and patients at risk of early relapse to be identified. Among node-positive patients, the prognostic value of the SPF value was confined to those with 1–3 positive axillary lymph nodes (the predominant node-positive subgroup), enabling a high and a low DRFS rate subgroup to be distinguished – a useful distinction where selection for adjuvant drug treatment is concerned. As the predictive strength of the SPF value was enhanced when correction was made for debris, we would recommend that the effect of such factors as debris be minimized as far as possible when flow cytometry-derived SPF values are to be used for prognostic purposes.