Lack of reliability of CA125 response criteria with anti-VEGF molecularly targeted therapy


The conclusion, rather definitively stated in the title (“Lack of reliability of CA125 response criteria with anti-VEGF molecularly targeted therapy”) of a recent Cancer article, must be challenged.1 The objection relates to the basic premise of the analysis. The investigators clearly state: “Response by RECIST imaging criteria was defined as the standard for response for this clinical study” (p. 1729). If the investigators had elected to define the ‘standard for response’ as the ‘Rustin CA-125 response criteria,’ an equally reasonable paradigm for monitoring therapeutic effect in ovarian cancer,2 then would the title of the article have been “Lack of reliability of RECIST imaging criteria with anti-VEGF molecularly targeted therapy”?

Although allowable space does not permit a thorough discussion of the serious deficiencies of radiographic imaging in evaluating response and progression in ovarian cancer, it is well documented that there are major disagreements between radiologists in their interpretation of abdominal CT scan findings in more than 25% of patients participating in antineoplastic drug trials in this malignancy.3 Furthermore, the recognized substantial biologic heterogeneity within the multiple metastatic deposits in ovarian cancer is a rational explanation for differences between a ‘measurable-mass–based’ view (radiographic imaging) versus what can be considered a more global ‘marker-based’ view of the status of the cancer in an individual patient.4

It certainly can be argued rationally that a modest rise in CA-125 in an otherwise ‘clinically stable’ ovarian cancer patient should not be the sole reason for discontinuation of the individual's current therapy. However, the identical argument can be made for radiographic evidence of a modest increase in the size of a measurable tumor mass.

One might have hoped the discussion and conclusion within, and title of, this article more accurately reflected the objective observation that there appear to be potentially clinically relevant inconsistencies between ‘CA-125’ and ‘RECIST imaging’ response criteria (Response Criteria in Solid Tumors) when molecularly targeted agents are administered in ovarian cancer. Furthermore, the existing data can be most interpreted most appropriately to raise at least as many questions regarding the utility of RECIST imaging criteria as they do for employing biologic markers of efficacy of molecularly targeted agents in this malignancy.

Maurie Markman MD*, * Department of Gynecologic Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas.