Progress in Defining Molecular Signature for Aggressive Androgen-Indifferent Prostate Cancer

Get Permission

Prostate cancer has lagged behind breast cancer in the identification of predictive and prognostic biomarkers, but the field is catching up. Researchers have identified a molecular signature that can distinguish aggressive prostate cancer that is androgen-indifferent and will have a better response to platinum-based therapy than to androgen receptor–based therapy. These aggressive variants include cancers with clinical features of small cell prostate cancer.

Prostate cancer is a heterogeneous disease, and there is even heterogeneity within risk groups. This work is a first step in defining clinically relevant subsets based on molecular characteristics.
— Ana Aparicio, MD

Tweet this quote

“We are interested in androgen-indifferent tumors—that is, tumors that don’t respond to androgen-directed therapy. These aggressive tumors do respond to platinum-based therapy, so our work is important in selecting patients for the right therapy,” explained Ana Aparicio, MD, of The University of Texas MD Anderson Cancer Center, Houston. “This discovery is critical, not only for our daily practice, but it should also make our clinical trials more effective by enriching patient selection for clinical trials.”

The molecular signature was identified in patients with castrate-resistant prostate cancer who have clinical features, but not morphologic features, of aggressive prostate cancer.

In routine practice, seven clinical features have been defined for aggressive variants of prostate cancer, Dr. Aparicio explained. They include predominantly lytic bone lesions, exclusive visceral metastases, low prostate-specific antigen (PSA) level relative to tumor burden, presence of neuroendocrine markers plus elevated carcinoembryonic antigen or lactate dehydrogenase level, < 6 months of response to androgen-deprivation therapy, bulky primary tumor > 5 cm, and small cell prostate cancer on biopsy.

“We have learned that patients may present with clinical features of small cell prostate cancer, but biopsy reveals a range of morphologies. We were interested in whether clinical features of aggressive prostate cancer alone can suggest response to therapy,” she continued. “When we studied this, we found that the answer is yes.”

Study Details

Dr. Aparicio and co-investigators conducted a phase II trial of patients with castrate-resistant prostate cancer and clinical features (as distinct from morphologic features) of small cell prostate cancer.1 “The study demonstrated that, in fact, if you selected men with clinical features of small cell prostate cancer, the response to platinum-based chemotherapy was high—as high as you would expect if patients had morphologic features of small cell prostate cancer,” she said.

Tailoring Treatment for Prostate Cancer

  • Researchers have identified a molecular signature for aggressive prostate cancer variants that are likely to be androgen-indifferent.
  • The signature was found in men with castrate-resistant prostate cancer.
  • The signature can be used to identify patients who should be treated with platinum-based therapy instead of androgen-directed therapy.

Next, they identified molecular features of small cell prostate cancer and then looked at clinically defined aggressive variants of prostate cancer to determine whether these variants had the same molecular signature. “The answer was yes,” she emphasized.

The molecular signature comprises genetic alterations in RB1, PTEN, and TP53—all related to tumor suppression. If a patient had two or more of these features, the cancer was defined as aggressive.

After validating the molecular signature in aggressive variants of prostate cancer, they conducted a trial of 160 unselected patients with metastatic prostate cancer randomized to cabazitaxel (Jevtana) vs cabazitaxel plus carboplatin (unpublished data). Patients were stratified for the presence or absence of features of the molecular signature. Again, they found that patients with the molecular signature (ie, abnormalities in at least two of the key genes) all had a good response to platinum-based therapy.

How Common Is Aggressive Prostate Cancer?

An older series of patients with castrate-resistant disease found small cell prostate cancer histology and gene expression in 8% of all tumors, combined genetic defects in 2 or more of these markers in 20%, and primary resistance to androgen receptor–signaling inhibitors in 20%. Among newly diagnosed patients, < 1% have small cell histology, 10% have combined genetic defects, and 10% have primary resistance to androgen receptor signaling.

“There is a spectrum of morphologic features that overlap with the genetic signature, and these cancers are platinum sensitive,” she emphasized.

She and her colleagues have mounted the ongoing DynaMO trial, which will enroll 265 men with castrate-resistant prostate cancer treated with abiraterone (Zytiga) and apalutamide (an investigational nonsteroidal antiandrogen) for 8 weeks. Serum markers will be measured at 8 weeks, and if there is a satisfactory decline in serum PSA level, patients will continue on chemotherapy with or without ipilimumab (Yervoy); if there is no satisfactory decline, patients will continue on abiraterone plus apalutamide with the addition of platinum-based chemotherapy with cabazitaxel and carboplatin. The expectation is that patients with an unsatisfactory response to the first 8 weeks of therapy will have aggressive cancers.

At the time of Dr. Aparicio’s presentation, 56 patients were accrued. In the first 35 evaluable patients, about one-third did not have a satisfactory decline in markers at 8 weeks and were allocated to chemotherapy.

“We believe that an understanding of genetic markers can categorize patients with prostate cancer into at least two major types: androgen-sensitive and androgen-indifferent. We have gone from morphologic features (ie, to define small cell prostate cancer) to clinical features that define aggressive variants, and now we have a molecular signature for these androgen-indifferent aggressive tumors. We hope that this will lead to improved outcomes in our patients,” she stated.

Translation to Clinical Practice

In routine clinical practice, oncologists can look for the seven criteria for aggressive variants of prostate cancer. Then the tumor can be sent out for genomic sequencing to look for alterations in RB1, PTEN, and/or TP53. These findings will inform treatment selection. Platinum-based therapy would be recommended for the androgen-indifferent tumors, she said.

“Prostate cancer is a heterogeneous disease, and there is even heterogeneity within risk groups. This work is a first step in defining clinically relevant subsets based on molecular characteristics,” Dr. Aparicio said.

The next study she and her colleagues are planning will enroll 160 patients with de novo metastatic disease and the presence of a bulky tumor (one of the clinical features of aggressive variants) to evaluate whether there is any benefit of removing the primary tumor surgically. Patients will get 6 months of systemic therapy, and then they will be randomized to undergo surgical excision of the primary tumor or not. “We will look for the molecular signature in these patients,” she noted. ■

Disclosure: Dr. Aparicio has received research funding from GlaxoSmithKline; advisory board honoraria from Sanofi; and clinical trial support from Janssen, Bristol-Myers Squibb, and Sanofi.


1. Aparicio A: Advances in the understanding and treatment of AR-indifferent disease. 2017 Genitourinary Cancers Symposium. General Session 2. Presented February 16, 2017.