Whole-Blood, 6-Gene Prognostic Signature May Help Personalize Prostate Cancer Treatment

William K. Oh, MD

Yixuan Gong, PhD

Prostate cancer is the most common cancer and second leading cause of cancer-related death in men in the United States. Though most patients with advanced disease initially respond to surgical or chemical depletion of serum testosterone, prostate cancer invariably progresses despite castrate levels of testosterone, a clinical state known as castration- resistant prostate cancer (CRPC). Ten to twenty percent of patients with prostate cancer develop CRPC within approximately 5 years of follow-up. CRPC is a strikingly heterogeneous disease state, and as a result, the overall survival of patients can range from a few months to many years. The ability to accurately predict prognosis in CRPC is critical for patient counseling and treatment decision making.

Traditionally, prognosis is based on clinical and laboratory variables, such as age, functional status, extent of bone and other metastases, prostate-specific antigen, alkaline phosphatase, and lactate dehydrogenase. More complicated nomograms have also been developed to combine these individual variables, but only offer moderate predictive power. A large number of immunohistochemistry-based prognostic tissue biomarkers have been proposed; however, a vast majority of these are not used in clinical practice, probably due to the lack of standardized methods to perform and interpret immunohistochemistry, the influence of tumor heterogeneity, and the lack of adequate tissues. There remains an urgent need to develop new prognostic models in CRPC.

Interactions between blood cells and the peripheral tissue through which blood circulates, including neoplastic tissue, might alter the gene expression of blood cells. Indeed, recent studies have shown that gene expression profiling of peripheral blood cells could yield diagnostic and prognostic information regarding various disease states. Expression profiling of whole blood offers several practical advantages compared with profiling tumor tissue, including the minimally invasive nature of sample acquisition, relative ease of standardization of sampling protocols, and the ability to obtain repeated samples over time.

We profiled the expression of 169 inflammation and prostate cancer—related target genes in whole blood of an initial cohort of 62 patients with CRPC. After using logistic regression and latent class methods to develop models, we were able to identify a 6-gene expression signature consisting of ABL2, SEMA4D, ITGAL, C1QA, TIMP1, and CDKN1A to separate men with low-risk and high-risk CRPC (7.8 months vs >34.9 months survival). A validation study of 140 additional patients confirmed these findings. Further analysis also showed that the prognostic ability of the 6-gene model was superior to clinicopathological variables (Lancet Oncol. 2012;13[11]:1105-1113).

The Leon and Norma Hess Center for Science and Medicine, Mount Sinai’s newest building, houses many of The Tisch Cancer Institute’s research and clinical facilities.

The successful validation of the whole-blood, 6-gene prognostic signature suggests that the host immune response is an important determinant of overall survival in patients with prostate cancer. We believe that this signature can be used not only as a prognostic marker, but also as criteria to stratify patients for some clinical trials. For example, one of the most exciting new developments in the treatment of patients with genitourinary cancers is the use of vaccines and immune-modulators, in particular with prostate and bladder cancers. At the Tisch Cancer Institute at Mount Sinai Medical Center, we have developed a program around sipuleucel-T, an immune treatment for prostate cancer. Our immune-based 6-gene signature may help to stratify patients and monitor the immune response to the therapy.

Our team is validating and refining the whole-blood prognostic signature using additional cohorts of patients with prostate cancer, conducting studies exploring the evolution of the signature during the course of a patient’s illness, and determining the predictive ability of this signature in patients with prostate cancer treated with immune-based therapies. Since it is possible that this signature is not unique to patients with prostate cancer, we are planning to look at the feasibility of the 6-gene signature in other types of malignancies. In the long run, we hope to validate the signature in larger validation cohorts and prospective trials, and eventually incorporate the signature into part of the personalized therapies for prostate cancer patients.