Year: 2005
Type: editorial
Abstract: The last decade has been marked by major advances in the field of biologically based therapies for solid tumors and the approval of several new therapeutic strategies. Despite the rational basis for these new treatments, the identification and validation of markers of prognostic and predictive value remains a challenge. Correlative data from studies with imatinib, trastuzumab, and epidermal growth factor receptor inhibitors suggest that the presence of the target does not guarantee a response to therapy. The successful determination of accurate prognostic and predictive markers hinges on establishing reproducible, sensitive methods of marker detection and unraveling the key components of the molecular pathways affected by a specific therapy. Angiogenesis is required for tumor growth and metastasis and is an attractive target for biologically based cancer therapy. It is a complicated, multistep process regulated by the balance of endogenous stimulatory and inhibitory factors elaborated by tumor cells and other cell types. Because of its seemingly pivotal role in tumorassociated angiogenesis across a wide range of malignancies, vascular endothelial growth factor-A (VEGF-A; commonly referred to as VEGF) has emerged as a central therapeutic target in cancer. VEGF is a potent, direct acting regulator of angiogenesis that signals through several receptor tyrosine kinases. Its expression is upregulated in virtually all types of cancer, and tumor-associated VEGF levels frequently correlate with microvascular density and predict for disease recurrence and decreased survival. Multiple isoforms are coexpressed in cancers, but the relative importance of individual splice variants remains ill defined. Bevacizumab (BV), a recombinant humanized monoclonal antibody directed against VEGF was approved by the US Food and Drug Administration in 2004 for use in patients with previously untreated metastatic colorectal cancer (mCRC). This was on the basis of data suggesting that the addition of BV to chemotherapy improves response rate, progression-free survival, and overall survival compared with chemotherapy alone. BV is generally well tolerated, but is associated with hypertension and small, but clinically significant, increases in GI perforations and arterial thromboembolic events. Recently, data from other phase III studies have further validated VEGF as a therapeutic target in cancer. Despite the overall impact of BV on survival in mCRC, the determinants of clinical benefit on an individual level remain undefined. Given the recent advances in our understanding of VEGF biology and the strong scientific rationale underlying this therapeutic strategy, it follows that the identification of valid prognostic and predictive markers for patients treated with BV should be feasible. In the accompanying article in this issue of the Journal of Clinical Oncology, Jubb et al explore the impact of VEGF expression, thrombospondin-2 expression and microvessel density (MVD) on the treatment effect of BV in metastatic colorectal cancer. The choice of potential markers reflects reports suggesting that 50% to 70% of colorectal cancers express VEGF, and the level of tumor-associated VEGF expression correlates with MVD and correlates inversely with prognosis. In addition to VEGF family members, a number of other regulators of angiogenesis (both positive and negative) are upregulated in colorectal cancer, although the relative importance of individual factors is unknown. Thrombospondin (THBS)-1 and THBS-2 have antiangiogenic activity, and both isoforms are expressed in human CRC. THBS-2 has been implicated in the inhibition of angiogenesis, even in the setting of high VEGF expression, and mRNA levels have been reported to correlate inversely with MVD and metastases. The authors performed a retrospective subset analysis using samples available from the 813 patients with untreated metastatic CRC (mCRC) randomly assigned to receive chemotherapy (irinotecan, fluorouracil, leucovorin [IFL]) with or without BV. Samples were available for approximately 35% of the patients, and 90% of the samples were from primary tumors. Epithelial and stromal VEGF expression were assessed at the level of protein (overall and maximal intensity) and mRNA on tissue microarrarys and whole sections. THBS-2 mRNA expression was scored in tumor-associated stroma. MVD was assessed in “hot spots” by immunohistochemistry as a continuous variable, as well as with varying cutoffs to define high and low MVD. Although the MVD score correlated with VEGF expression, none of the markers proved to be significant prognostic or predictive factors. Thus, the addition of BV to IFL in patients with mCRC improves survival regardless of the level of VEGF expression, THBS-2 mRNA expression, or MVD; none of the markers discriminate between patients more or less likely to benefit from the addition of BV to first-line chemotherapy. Assuming that the characteristics of the subgroup reflected those of the group as a whole, a larger sample size would probably not have changed the conclusion. At first glance, these findings appear to be at odds with previously published data related to prognostic markers in CRC and the concept that the presence of the target should predict response to therapy. However, these results are consistent with data generated JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 24 NUMBER 2 JANUARY 1
Source: Journal of Clinical Oncology
Author: Emily K. Bergsland
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: bronze
OpenAlex ID: W2120317878