A novel machine learning-derived decision tree including uPA/PAI-1 for breast cancer care.

Background uPA and PAI-1 are breast cancer biomarkers that evaluate the benefit of chemotherapy (CT) for HER2-negative, estrogen receptor-positive, low or intermediate grade patients. Our objectives were to observe clinical routine use of uPA/PAI-1 and to build a new therapeutic decision tree integrating uPA/PAI-1. Methods We observed the concordance between CT indications proposed by a canonical decision tree representative of French practices (not including uPA/PAI-1) and actual CT prescriptions decided by a medical board which included uPA/PAI-1. We used a method of machine learning for the analysis of concordant and non-concordant CT prescriptions to generate a novel scheme for CT indications. Results We observed a concordance rate of 71% between indications proposed by the canonical decision tree and actual prescriptions. Discrepancies were due to CT contraindications, high tumor grade and uPA/PAI-1 level. Altogether, uPA/PAI-1 were a decisive factor for the final decision in 17% of cases by avoiding CT prescription in two-thirds of cases and inducing CT in other cases. Remarkably, we noted that in routine practice, elevated uPA/PAI-1 levels seem not to be considered as a sufficient indication for CT for N≤3, Ki 67≤30% tumors, but are considered in association with at least one additional marker such as Ki 67>14%, vascular invasion and ER-H score <150. Conclusions This study highlights that in the routine clinical practice uPA/PAI-1 are never used as the sole indication for CT. Combined with other routinely used biomarkers, uPA/PAI-1 present an added value to orientate the therapeutic choice.

Assessment of Ki67 and uPA/PAI-1 expression in intermediate-risk early stage breast cancers.

BACKGROUND: The objective of this study was to compare the efficacy of biomarkers in assessing the risk of breast cancer recurrence in patients with node-negative or micrometastatic grade II breast cancer. Specifically, we compared risk assessments based on the St. Gallen clinicopathological criteria, Ki67 expression and urokinase plasminogen activator (uPA)/plasminogen activator inhibitor-1 (PAI-1) expression. METHODS: This retrospective study included 347 patients with breast cancer followed at Limoges University Hospital. The optimal cut-off for high Ki67 expression (Ki67hi) was established as 20%. The threshold for uPA and PAI-1 positivity was 3 ng/mg and 14 ng/mg, respectively. RESULTS: Ki67 expression was lower in uPA/PAI-1-negative than in uPA/PAI-1-positive tumours (227 tumours; P = 0.04). The addition of Ki67 status to the St. Gallen criteria resulted in a 28% increase in the rate of identification of high-risk tumours with a potential indication for chemotherapy (P < 0.001). When considering uPA/PAI-1 levels together with the St Gallen criteria (including Ki67 expression), the number of cases identified as having a high recurrence risk with a potential indication for adjuvant chemotherapy increased by 20% (P < 0.001). Adjuvant chemotherapy was 9% less likely to be recommended by a multidisciplinary board when using the current criteria compared with using a combination of the St. Gallen criteria and Ki67 and uPA/PAI-1 status (P = 0.03). CONCLUSIONS: Taken together, our data show discordance among markers in identifying the risk of recurrence, even though each marker may prove to be independently valid.

European cost-effectiveness study of uPA/PAI-1 biomarkers to guide adjuvant chemotherapy decisions in breast cancer.

BACKGROUND: This study investigated the cost effectiveness of guideline-recommended (American Society of Clinical Oncology, European Society of Medical Oncology) urokinase plasminogen activator (uPA)/plasminogen activator inhibitor-1 (PAI-1) biomarkers to guide adjuvant chemotherapy decisions for hormone receptor-positive, node-negative early breast cancer patients at intermediate risk of relapse, in France, Germany, and The Netherlands. METHODS: uPA/PAI-1 testing was compared to chemotherapy for all patients and to no chemotherapy in two age-related subgroups (35-49 and 50-75 years). A partitioned survival analysis was performed using patient-level data for survival outcomes and secondary sources. Mean quality-adjusted life years (QALYs) and costs were estimated over a lifetime horizon to calculate the incremental net monetary benefit (INMB) at a willingness-to-pay of €50,000/QALY. Uncertainty was explored through bootstrap and probabilistic sensitivity analysis using 5000 replicates. RESULTS: In the 35-49 year age group, INMBs were negative when uPA/PAI-1 testing was compared to chemotherapy for all patients but positive when it was compared to no chemotherapy for the three countries. In the 50-75 year age group, INMBs of uPA/PAI-1 testing compared to both reference strategies were positive in the three countries, with cost-effectiveness probabilities for the uPA/PAI-1 strategy of 65%, 70%, and 59% for France, Germany, and the Netherlands, respectively, compared with chemotherapy for all patients, and 64%, 58%, and 65%, respectively, compared with no chemotherapy. CONCLUSIONS: uPA/PAI-1 testing could allow the selection of patients older than 50 years requiring chemotherapy in this population, but the cost effectiveness of this strategy is uncertain. Chemotherapy for all patients is the most cost-effective strategy for patients younger than 50 years.