AACR 2009: Molecular Profiling of Tumors Improves Cancer Treatment

Zosia Chustecka

April 21, 2009

April 21, 2009 (Denver, Colorado) — A pilot study of molecular profiling of tumors, conducted in patients with advanced cancers that were progressing despite several previous treatments, found that the process helped identify therapies that ultimately had an impact on the disease. Shrinkage of tumors was shown in 47% of patients. Also, 27% of the profiled patients showed improvement in progression-free survival, compared with that seen with the previous therapy; there is a suggestion of improved overall survival in these patients.

"We consider this to be a promising result," said lead researcher Daniel Von Hoff, MD, physician-in-chief at TGen, a nonprofit research institute in Phoenix, Arizona, and chief scientific officer at Scottsdale Healthcare and US Oncology Research.

"With this trial, we are showing the power of personalized medicine using tools we have already available to us," he commented.

Presenting the results at a plenary session here during the American Association for Cancer Research 100th Annual Meeting, Dr. Von Hoff acknowledged that the study was small (66 patients) and that, because these patients acted as their own controls, a larger randomized trail is needed. But the results demonstrate clinical benefits from molecular profiling, and also show that better treatment can be achieved using existing technology and drug therapies.

"This is a very strong step in the right direction," said discussant Ronald DePinho, MD, from the Dana-Farber Cancer Institute, in Boston, Massachusetts, who is also a professor of medicine at Harvard Medical School. "This is very encouraging. These are obviously preliminary results, but this is what we absolutely need to do — profile a patient's tumor and then use personalized therapy."

"We need to do more of this," Dr. DePinho added.

Profiling Identified a Precise Target

In this study, tumors were evaluated with immunohistochemistry and microassay profiling to identify precise targets, and treatment regimens were based on these targets. In many of these refractory tumors, targets for conventional therapies were identified, which was "a surprise finding," Dr. Von Hoff commented. But the profiling also suggested therapies in cases where the treating physician was unsure what to use next; in the case of an eccrine sweat-gland tumor, which profiling suggested would respond to sunitinib, it did, he added.

Of the 66 participants, 27% had breast cancer, 17% had colorectal cancer, and 8% had ovarian cancer; the remainder were classified as miscellaneous.

Patients acted as their own controls, which eliminated confounding factors, such as tumor subtypes and differences in biology, Dr. Von Hoff said. "We compared progression-free survival, after the patients progressed on their prior treatment regimen, with what had occurred after molecular profiling," he explained.

Overall, 18 of the 66 patients (27%) showed a significant improvement in progression-free survival (P = .007). "It was also encouraging to see that the overall survival in these 18 patients was better than that for the whole group of 66 patients (9.7 vs 5 months)," he said. The improvement in progression-free survival was 44% in patients with breast cancer, 36% in those with colorectal cancer, 20% in those with ovarian cancer, and 16% in the miscellaneous group.

Already Used in Clinical Practice at Leading Institutions

Although this is a pilot study, molecular profiling is already being used in clinical practice at several leading cancer institutions, including Massachusetts General Hospital (MGH), in Boston.

"We are already using molecular profiling for all our lung cancer patients," Jeffrey Settleman, PhD, scientific director at the MGH Cancer Center, told Medscape Oncology. This has already had an impact on treatment decisions, and it appears to be improving treatment. "We have seen better response rates and we hope that this will translate into better survival," he commented.

"Within the next year, we plan to use molecular profiling on all cancer patients, who number about 5000 annually," he added. The MGH Cancer Centre uses a PCR-based mutation-detection assay and state-of-the-art robotic technology, called SNaPshot, to look for 110 known gene mutations in tumor tissue.

The technology used in molecular profiling of tumors is not proprietary, and several other institutions have developed their own systems, including the University of Texas MD Anderson Cancer Center and the Memorial Sloan-Kettering Cancer Center, Dr. Settleman said. All are striving to profile individual tumors so that therapy can be personalized, which means that it has a better chance of working because it targets specific mutations found in that tumor. This also prevents patients from being exposed to drugs that have a limited chance of success, eliminating toxicity and improving quality of life, he noted.

The use of targeted therapies, which are very expensive, will become more efficient, Dr. Settleman predicted, because they will be used only in patients who are likely to see benefit.

However, the promise of such personalized therapy is restricted by the small number of targeted therapies that is currently available, he said.

This point was raised by Dr. DePinho is his discussion. "There is a meager portfolio of cancer drugs, and many of the products in [research and development] are directed against the same pathways," he commented. "We need to increase our repertoire, and we need a dramatic expansion in the cancer-drug portfolio."

This will come, Dr. DePinho predicted, as ongoing efforts uncover epigenetic alterations in the human cancer. "So far we know only about 5% of the genes that are involved in human cancer."

As more mutations are found and new drugs targeting them are developed, this field of personalized medicine will expand and will lead to more effective treatment of cancer, probably within the next half-dozen or dozen years, he said.

The study was funded by grants from the Stardust and Scottsdale Healthcare Foundations. The researchers have disclosed no relevant financial relationships.

American Association for Cancer Research (AACR) 100th Annual Meeting: Abstract ALB-259. Presented April 20, 2009.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....