Boston, MA—Biomarker development and validation are essential for the rational use of emerging cancer treatments, said presenters at the second Global Biomarkers Consortium annual conference.
“If we don’t identify biomarkers, we’re likely to miss a therapeutic effect,” said Rob Coleman, MD, MBBS, Director, Sheffield Cancer Research Centre, United Kingdom. “We want to separate our patients into those that express the biomarker and those that do not, with the expectation that those that do will be enriched, and it will be easier to show benefit.”
The 3 phases in the development of a clinically useful biomarker are discovery, validation, and application. Clinical and biological validity of a potential biomarker is common, analytic validity is less common, and proving clinical utility, whether prognostic or predictive, is rare, said Dr Coleman.
“To be accepted as clinically useful, a biomarker needs to have gone through prospective clinical trial testing, or at least have been subjected to a meta-analysis,” he said. “Relatively few achieve that.”
Showing the value of a new biomarker requires complete publication of the methodology used in its evaluation, according to the Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK) criteria. These include patient characteristics; the type of biologic material used and its collection, preservation, and storage; study design and case selection; a list of candidate variables examined or considered; and the handling of marker values and cut point determination. The REMARK criteria are among several guidelines that have been established for reporting and evaluating biomarker studies.
Biomarker trial design relies on the availability of adequate archived specimens for analysis to have acceptable statistical power, said Dr Coleman. The marker-based test should be analytically and preanalytically validated for use with the archived specimens.
Searching for Bone Metastases Marker in Breast Cancer The use of adjuvant bisphosphonates in early breast cancer is being studied in more than 3000 patients with metastatic breast cancer in the randomized Adjuvant Zoledronic Acid to Reduce Recurrence (AZURE) study, with the rationale that such treatment may enhance endocrine therapy and prevent metastasis. Although disease-free survival was similar between the zoledronic acid (Zometa) arm and the controls, there was a 23% reduction (P <.05) in invasive disease events in postmenopausal women who received the bisphosphonate. The agent may be working to prevent bone metastases, but to use the bisphosphonate in the most appropriate and cost-effective manner, “you need to pull out a group of patients who are more likely to develop bone metastases,” he said, “and there’s no tool for doing that. We don’t have any biomarkers that do this.”
As an example of biomarker discovery and validation, his group has used a proteomic discovery platform to identify 2 candidate proteins for the prediction of bone metastases, which are being evaluated using tissue microarray samples from the AZURE study population.
When both proteins were expressed (in 15% of the study patients), the risk for bone metastases plus metastases at other sites was increased by more than 3-fold in the control group, and the risk of bone metastases only was increased by more than 4-fold in the controls. No such increase was found in the groups that were randomized to the bisphosphonate.
Markers in Hematologic Malignancies In the case of biomarkers for hematologic malignancies, chronic myelogenous leukemia (CML) is characterized by the Philadelphia chromosome–positive (Ph+) status, creating the fusion gene BCR-ABL. The Ph+ chromosome is not only pathogenic for CML but is also a tumor marker. The tyrosine kinase inhibitors (TKIs) block excess tyrosine kinase activity caused by the BCR-ABL gene, changing the natural history of the disease.
“As good as the treatment is, the testing for the presence of molecular abnormalities at the BCR level is poorly homogenized,” said Jorgé E. Cortes, MD, Deputy Department Chair, Department of Leukemia, M.D. Anderson Cancer Center, Houston. Standardization of the test is lacking, resulting in large differences between laboratories in the measurement of the number of leukemic cells.
An undetectable molecular abnormality achieved with targeted therapy has been associated with excellent long-term survival. Not every BCR-ABL rearrangement is the same, however, and the response to therapy is different between patients with typical and atypical rearrangements, said Dr Cortes. The e1a2 transcript is rare in patients with CML, and patients with this transcript have inferior outcomes compared with those with more common transcripts.
Another marker of response in CML is the human organic cation transporter (hOCT1). Patients with an active hOCT1 experience cytogenetic and molecular responses to a standard dose of a TKI, whereas those with a less active transporter require higher doses of the drug.
In patients with CML, BCR-ABL can mutate to confer resistance to a TKI. “There are more than 100 mutations that can occur, and they create different levels of resistance to imatinib [Gleevec],” Dr Cortes said. Second- and third-generation TKIs have been developed to overcome resistance caused by specific mutations, “and they overcome some mutations better than others,” he said. “Knowing which mutation is more likely to respond to which drug helps you individualize treatment for that patient.”
Patients with CML and the T315I mutation currently have no treatment options. Ponatinib (Iclusig) is a potent pan–BCR-ABL inhibitor that is active against all tested resistant mutations, including the T315I mutation. In a phase 1 study of 403 patients resistant to or intolerant of dasatinib (Sprycel) or nilotinib (Tasigna), or those who had the T315I mutation, early complete cytogenetic responses were achieved in 58% of patients who received ponatinib, said Dr Cortes.
In acute myelogenous leukemia (AML), a mutation in FLT3 is associated with frequent and rapid relapse and worse overall survival. In a retrospective study of patients with myelodysplastic syndrome or AML, 23 patients received an FLT3 inhibitor as part of their induction, and 9 achieved a complete remission or a complete remission with incomplete platelet recovery. A secondary tyrosine kinase domain mutation can arise after the use of FLT3 inhibitors in patients with single FLT3-internal tandem duplication–mutated AML, conferring resistance and a poor prognosis.
A molecular mechanism of AML is the process of farnesylation, which is required for the activation of reninangiotensin system proteins. The farnesyltransferase inhibitor tipifarnib (Zarnestra) has been evaluated in untreated elderly patients with AML, producing a complete remission in 18% to 20%.
A genetic predictor of response to tipifarnib is the ratio of RASGRP1/APTX gene expression. The utility of this classifier for predicting response to tipifarnib was validated in a set of 58 samples from patients with relapsed or refractory AML, and was found to predict improved survival in patients with newly diagnosed and refractory or relapsed AML.
