New Appproach to Clinical Trials; Emphasis on Individual Patient Response
Posted Oct 05 2010 12:00am
Concern has been building for a number of years about the way that clinical trials for cancer drugs were being conducted. An alternative method is now being tested. It provides far more benefits for the enrolled patients and could be a model for future trials if it is successful. Details were provided in a recent article (see: A New Rx for Medicine ; subscription required). Below is an excerpt from it:
[A novel clinical trial for breast cancer], called I-Spy 2 , employs several innovative approaches to improve the notoriously slow and inefficient process of developing new cancer drugs. Using the latest advances in genetics, I-Spy 2 aims to match experimental drugs with the molecular makeup of tumors most likely to respond to them. And it tests multiple drugs at once, with the intent of getting the most effective ones into late-stage trials more quickly. The goal is to reverse a dismal record, in which 60% to 70% of late-stage cancer studies fail, and to dramatically reduce the time and cost required to get promising new drugs to the market....The new trial draws from a growing body of research showing that the genetic makeup of tumors varies widely even among patients diagnosed with the same cancer. Some of these molecular traits, called biomarkers, make a tumor vulnerable to a specific medicine, while others may thwart a drug's effect. A big part of the problem with conventional trials is that they essentially take all comers....The study focuses on women with aggressive breast cancers that have not spread to other organs. It aims to collect information about experimental drugs that would then enable drug companies to design leaner, faster late-stage trials that enroll only patients whose tumors had a high probability of responding to the treatment...."The vision is a 300-patient phase III trial instead of a 3,000-patient trial, with better results," says [a] co-leader of the study....The trial incorporates other new approaches. It's currently testing five drugs at once from three different companies. As compounds graduate to phase III studies or are winnowed out for lack of significant benefit, new candidates are cycled in, sparing the cost and time to mount separate trials. Up to 12 candidates will be screened over the course of the study....In addition, participants get a six-month course of chemotherapy before having an operation to remove their tumors. In many studies, surgery is performed first, followed by chemotherapy and radiation, reflecting a long-standing preference to quickly remove tumors before beginning drug treatment. As a result, it can take three to five years to determine whether a drug is working, as researchers wait to see if the cancer recurs. Whether surgery occurs before or after chemo does not affect long-term outcomes. In I-Spy 2, researchers use an MRI to evaluate a tumor's response to a drug early in the trial and get a definitive answer at surgery. "All of a sudden I've taken a five-year learning curve and shortened it to six months," [said a spokesman].
As noted above and to summarize, I-Spy 2 is unique among other clinical trials on a number of counts: (1) five experimental drug are currently being tested with additional candidate drugs being cycled into the study as necessary; (2) each research subject is initially matched to a particular drug on the basis of the biomarkers expressed by the malignant lesion, thus providing early evidence of a desirable therapeutic effect; (3) patients entering the study are not subjected to immediate surgery but rather treated initially with chemo with the effectiveness of the drug in terms of tumor shrinkage measured with MRI. From the perspective of the study subjects, this is an ideal situation because drug treatment is customized for each of them based on measurable effects. Compare this with previous study design in which a new treatment is compared against conventional therapy. Those assigned to the latter group were generally no better off personally despite their participation in the study.
For me, the first idea that came to mind when I read this story was that finally our most sophisticated diagnostic methods (i.e., tissue biomarkers and MRI) were being utilized to design "personalized" therapy for patients in clinical trials. The second was that such a novel approach might require new biostatistical techniques to analyze the findings. However, given the fact that the therapeutic effects are being measured using a quantitative tool, medical imaging, the biostatistical analysis might not be challenging as I initially imagined. At the very least and from where I sit, this new approach to clinical trials is a major advance for the enrolled patients.
Concern has been building for a number of years about the way that clinical trials for cancer drugs were being conducted. An alternative method is now being tested. It provides far more benefits for the enrolled patients and could be a model for future trials if it is successful. Details were provided in a recent article (see: A New Rx for Medicine ; subscription required). Below is an excerpt from it:
[A novel clinical trial for breast cancer], called I-Spy 2 , employs several innovative approaches to improve the notoriously slow and inefficient process of developing new cancer drugs. Using the latest advances in genetics, I-Spy 2 aims to match experimental drugs with the molecular makeup of tumors most likely to respond to them. And it tests multiple drugs at once, with the intent of getting the most effective ones into late-stage trials more quickly. The goal is to reverse a dismal record, in which 60% to 70% of late-stage cancer studies fail, and to dramatically reduce the time and cost required to get promising new drugs to the market....The new trial draws from a growing body of research showing that the genetic makeup of tumors varies widely even among patients diagnosed with the same cancer. Some of these molecular traits, called biomarkers, make a tumor vulnerable to a specific medicine, while others may thwart a drug's effect. A big part of the problem with conventional trials is that they essentially take all comers....The study focuses on women with aggressive breast cancers that have not spread to other organs. It aims to collect information about experimental drugs that would then enable drug companies to design leaner, faster late-stage trials that enroll only patients whose tumors had a high probability of responding to the treatment...."The vision is a 300-patient phase III trial instead of a 3,000-patient trial, with better results," says [a] co-leader of the study....The trial incorporates other new approaches. It's currently testing five drugs at once from three different companies. As compounds graduate to phase III studies or are winnowed out for lack of significant benefit, new candidates are cycled in, sparing the cost and time to mount separate trials. Up to 12 candidates will be screened over the course of the study....In addition, participants get a six-month course of chemotherapy before having an operation to remove their tumors. In many studies, surgery is performed first, followed by chemotherapy and radiation, reflecting a long-standing preference to quickly remove tumors before beginning drug treatment. As a result, it can take three to five years to determine whether a drug is working, as researchers wait to see if the cancer recurs. Whether surgery occurs before or after chemo does not affect long-term outcomes. In I-Spy 2, researchers use an MRI to evaluate a tumor's response to a drug early in the trial and get a definitive answer at surgery. "All of a sudden I've taken a five-year learning curve and shortened it to six months," [said a spokesman].As noted above and to summarize, I-Spy 2 is unique among other clinical trials on a number of counts: (1) five experimental drug are currently being tested with additional candidate drugs being cycled into the study as necessary; (2) each research subject is initially matched to a particular drug on the basis of the biomarkers expressed by the malignant lesion, thus providing early evidence of a desirable therapeutic effect; (3) patients entering the study are not subjected to immediate surgery but rather treated initially with chemo with the effectiveness of the drug in terms of tumor shrinkage measured with MRI. From the perspective of the study subjects, this is an ideal situation because drug treatment is customized for each of them based on measurable effects. Compare this with previous study design in which a new treatment is compared against conventional therapy. Those assigned to the latter group were generally no better off personally despite their participation in the study.
For me, the first idea that came to mind when I read this story was that finally our most sophisticated diagnostic methods (i.e., tissue biomarkers and MRI) were being utilized to design "personalized" therapy for patients in clinical trials. The second was that such a novel approach might require new biostatistical techniques to analyze the findings. However, given the fact that the therapeutic effects are being measured using a quantitative tool, medical imaging, the biostatistical analysis might not be challenging as I initially imagined. At the very least and from where I sit, this new approach to clinical trials is a major advance for the enrolled patients.