uPA/PAI-1 Predict Adjuvant Chemo Benefit in Breast Cancer

SAN ANTONIO—In women with primary breast cancer, the tumor invasion factors urokinase-type plasminogen activator (uPA) and plasminogen activator inhibitor type 1 (PAI-1) can predict benefit from adjuvant chemotherapy and should support clinical decision making, according to European investigators who for years have been studying the role of these two molecules in breast cancer. Nadia Harbeck, MD, of Technical University, Munich, Germany, described their most recent findings at the 25th Annual San Antonio Breast Cancer Symposium (abstract 7).

The two factors play a key role in tumor invasion, performing actions that are critical for the metastatic phenotype. They were the first novel biological factors to demonstrate solid clinical relevance in breast cancer by affecting prognosis. This was shown in a prospective randomized trial by the European Organization for Research and Treatment of Cancer (EORTC) and in a pooled analysis of more than 8,000 EORTC patients.

"The evidence and the availability of quality-controlled assays prompted us to include these factors among those we use to make clinical decisions in node-negative patients," Dr. Harbeck said. "We saw that these factors clearly differentiate the prognosis in patients who did not receive adjuvant therapy, either node-negative or node-positive patients." In patients who did receive adjuvant therapy in earlier analyses, she said, the differences between patients with low vs high levels of uPA and PAI-1 were diminished, and there was little prognostic impact.

In a prospective multicenter trial of node-negative patients, those considered at high risk by uPA and PAI-1 levels were randomized to observation (after loco-regional treatment) or chemotherapy. Results showed that chemotherapy reduced the risk of relapse in these patients. The high-risk patients who received no adjuvant chemotherapy had a significantly worse prognosis than the low-risk patients. Their risk was comparable to that of patients with three to four positive lymph nodes (Harbeck N et al: San Antonio Breast Cancer Symposium 2001, abstract 19).

"These were hints, but they did not prove that these factors were actually predictive, only that high-risk patients benefit from chemotherapy," she said.

The Current Study

The current study, therefore, was designed to determine whether relapses that would occur (predicted with conventional decision-making factors) could be avoided or delayed by measuring these invasion factors and administering chemotherapy to the high-risk group.

The study was based on a combined dataset of 3,424 primary breast cancer patients from the Technical University of Munich and the Rotterdam Cancer Institute: uPA and PAI-1 levels were measured by ELISA in tumor tissue extracts, and a binary variable uPA/PAI-1 (either test high vs both tests low) was used.

After a median follow-up of 83 months, uPA/PAI-1 was shown to significantly affect disease-free survival in a Cox multivariate analysis; the hazard ratio for recurrence was 2.0 for patients with high levels (P < .001), she reported. "The key observation is that patients with high uPA/PAI-1 have an enhanced benefit from adjuvant chemotherapy, compared to those with low levels," she said.

This effect is seen as a significant interaction between chemotherapy and uPA/PAI-1 for the entire group (P < .003; hazard ratio = 0.68) and separately within nodal subgroups (0 to 3 and greater than 4 involved nodes), she reported.

In patients with 0 to 3 nodes, the benefit from chemotherapy is significantly greater for patients with high uPA/PAI-1 values than for patients with low values. There was also a marked benefit from chemotherapy in patients with high uPA/PAI-1 levels and more than 4 positive nodes.

The differences exerted by low vs high uPA/PAI-1 levels were less evident in patients who received hormonal therapy. "With chemotherapy, there was an especially high benefit in patients with high uPA/PAI-1, but for hormonal treatment, it didn’t matter if the patient’s uPA/PAI-1 levels were high or low; both groups benefited," she pointed out.

The investigators concluded that uPA/PAI-1 do have a prognostic affect in breast cancer. "Node-negative patients with low uPA/PAI-1 levels have a very high probability of being cured by loco-regional therapy alone and have no need for adjuvant chemotherapy. But node-negative patients with high uPA/PAI-1 are at high risk for relapse, much higher than you would expect, and they need aggressive adjuvant systemic therapy," she said.

Dr. Harbeck cautioned that the results "should not be interpreted as evidence for removing endocrine therapy where indicated but, rather, for measuring uPA and PAI-1, and—in light of the known additive benefits of endocrine therapy and chemotherapy—for administering both adjuvant endocrine therapy and adjuvant chemotherapy to the high-risk group." The study has been published in Cancer Research (62:4617-4622, 2002).