Gene Expression-Based Prognostic and Predictive Markers for Breast Cancer: A Primer for Practicing Pathologists by ProQuest


More Info
									        Gene Expression–Based Prognostic and Predictive
                  Markers for Breast Cancer
                                       A Primer for Practicing Pathologists
                                 Chungyeul Kim, MD; Yusuke Taniyama, MD; Soonmyung Paik, MD

● Context.—Gene expression–based prognostic assays for                Project. This review focused on a general conceptual de-
breast cancer are now available as commercial reference               scription of the technology behind these assays and differ-
laboratory tests covered by insurance.                                ences among them to aid understanding by pathologists in
  Objective.—To provide practicing pathologists with in-              practice.
formation about the nature of these assays, differences                 Conclusions.—While these assays are clinically useful,
among them, and their use by clinical oncologists in the              they are still evolving. The future development of gene ex-
management of patients diagnosed with breast cancer.                  pression–based markers will need to be more clinical-con-
  Data Sources.—Review of literature and unpublished                  text–specific to be clinically useful.
data from the National Surgical Adjuvant Breast and Bowel               (Arch Pathol Lab Med. 2009;133:855–859)

                  CLINICAL BACKGROUND                                 ment can consist of surgery alone, surgery plus antihor-
   The identification of molecular markers with prognostic             monal therapy, or surgery plus chemohormonal therapy,
significance may help cancer patients avoid treatment that             depending on the stage and hormone receptor status of
is unlikely to be successful. In breast cancer, for example,          the tumor. This varying clinical context dictates that most
clinical studies have shown that adding adjuvant chemo-               clinically useful prognostic and predictive markers are
therapy to tamoxifen in the treatment of node-negative,               those developed with a specific clinical context in mind
hormone receptor (HR)–positive breast cancer improves                 and tested and validated within that clinical context. Clin-
disease outcome.1 However, treatment with tamoxifen                   ical tests can provide either prognostic or predictive in-
alone is associated with a 15% likelihood of distant recur-           formation or both.
rence at 10 years in this population, suggesting that 85%                In the past there was no distinction made between
of these patients would do well without the addition of               ‘‘prognostic’’ and ‘‘predictive’’ markers when investigators
cytotoxic chemotherapy and could avoid the adverse                    conducted studies and such practice was a source of great
events inherent to such treatment.1 Nevertheless, the cur-            confusion in the field. More recently, a distinction between
rent National Comprehensive Cancer Network (NCCN)2                    the two has become the norm rather than the exception.
and St. Gallen3 clinical practice guidelines, using classical         However, it also has become clear that for most chemo-
histopathology and immunohistochemical prognostic                     therapy regimens in use today, general prognosticators
markers, categorize less than 10% of patients with node-              end up as predictive markers of the degree of benefit from
negative, HR-positive disease at low enough risk of recur-            chemotherapy. For example, estrogen receptor status is not
rence to forgo adjuvant chemotherapy. These treatment                 only prognostic and predictive of response to antihor-
guidelines assume that patients will derive the same de-              mone therapy, but also predictive of response to chemo-
gree of benefit from chemotherapy regardless of their                  therapy (the presence of the estrogen receptor is associ-
baseline risk.                                                        ated with less benefit from chemotherapy). Another ex-
   Clinical oncologists need 2 basic pieces of information            ample is uPA/PAI-1 (urokinase-type plasminogen activa-
to make treatment decisions for individual patients: (1) a            tor/plasminogen activator inhibitor type 1). Harbeck and
patient’s baseline risk after legacy treatment and (2) the            colleagues4 analyzed samples from more than 8000 wom-
expected degree of additional benefit she will receive from            en with node-negative breast cancer. Intratumoral concen-
systemic therapy added to legacy treatment. Legacy treat-             tration of uPA and its inhibitor, PAI-1, were found to cor-
                                                                      relate directly with risk of disease recurrence. Despite the
                                                                      fact that these markers have no suspected direct role in
  Accepted for publication November 10, 2008.                         chemotherapy response, patients categorized as high-risk,
  From the Division of Pathology, National Surgical Adjuvant Breast   based on uPA/PAI-1 levels, derived significant benefit
and Bowel Project Foundation, Pittsburgh, Pennsylvania.               from chemotherapy, whereas low-risk patients gained lit-
  The authors have no relevant financial interest in the products or
To top