Cognitive impairment is increasingly recognized as an important problem in schizophrenia. Cognitive deficits are considered to be a core feature of schizophrenia (e.g., not related to medications or psychotic symptoms) and they are determinants of how well patients function in the community. This paper was the first publication from an initiative sponsored by the National Institute of Mental Health (NIMH) that was designed to stimulate pharmacological treatments for cognitive impairment in schizophrenia called "Measurement and Treatment Research to Improve Cognition in Schizophrenia" (MATRICS).

Not exactly—it does something different. As of a few years ago, there was no pathway by which a drug could be approved by the US Food and Drug Administration (FDA) for cognitive impairment in schizophrenia. Hence, the pharmaceutical industry had limited interest in the area because there was no avenue for them to market new drugs in this area. The goals of MATRICS were to reach consensus on key points that were necessary before the FDA would grant approval for drugs in this area.

For example, consensus was needed on the endpoint for clinical trials, including a standard definition and measurement of cognitive performance. Consensus was also needed on design elements of clinical trials in this area. MATRICS involved several hundred scientists from academia, industry, NIMH, and the FDA to reach agreement on these topics. This paper describes the rationale and the multi-step process that was used to arrive at a consensus cognitive battery for use in clinical trials of cognition enhancement in schizophrenia.

The paper describes a necessary step for developing drugs for a new clinical target. The goal is to build a pathway for FDA approval so that pharmaceutical companies can develop and market new types of drugs for cognitive impairment in schizophrenia. The first step in this process was to decide how to define and measure cognitive impairment in schizophrenia. This paper explained why cognition in schizophrenia is an important clinical target for intervention, how cognition-enhancing drugs may improve the functioning of patients, and it outlined the steps that would be taken to arrive at a consensus measure for cognition.

MATRICS was a competitive NIMH contract. A group from UCLA coordinated an application (involving scientists from several universities) and we received the contract. I served as co-Principal Investigator on the MATRICS Initiative and Stephen Marder was Principal Investigator. Keith Nuechterlein and I were co-chairs of the MATRICS Neurocognition Committee, which oversaw the process for selecting a consensus battery.

We encountered numerous conceptual, organizational, and practical challenges over the three years of MATRICS. There were no obvious models from previous initiatives that we could borrow and use for guidance. One challenge was to decide on the cognitive domains that should be assessed in a consensus battery, a step that required a careful review of the literature and discussion at a public meeting.

Other decisions required new data collection on the psychometric properties of candidate tests. One fundamental challenge was that pharmaceutical companies were already invested in certain cognitive tests that they had used previously, and test developers were naturally invested in their own tests. It was a situation in which everyone needed to be convinced to give a little so that everyone would benefit and we could select a consensus measure that would enable FDA to approve new drugs in this area.

There are several follow up initiatives from MATRICS. One is a clinical trial network sponsored by NIMH called "Treatment Units for Neurocognition and Schizophrenia" (TURNS). This network of seven sites is using the products of MATRICS and is currently testing promising new compounds for cognition enhancement in schizophrenia.

Another NIMH initiative is called "Cognitive Neuroscience Treatment Research to Improve Cognition in Schizophrenia" (CNTRICS). It was noted that tests that come from cognitive neuroscience often have excellent construct validity, but psychometrics that are less than optimal for clinical trials. The CNTRICS Initiative will consider ways in which promising measures from cognitive neuroscience can be modified so they obtain the psychometric properties needed for clinical trials.

Another follow up activity is MATRICS–CT (Co-primary and Translation), which was initiated by the pharmaceutical industry. It includes participants from academia and NIMH, but is funded entirely by industry. MATRICS-CT has two goals: One goal is to evaluate potential "co-primary" measures which are functionally meaningful outcome measures that can be used in clinical trials along with cognitive performance measures. The other goal is to translate the consensus cognitive battery from MATRICS into foreign languages to facilitate international clinical trials.

Yes, the results of this NIMH initiative have direct public health significance. The goal is to treat a part of schizophrenia (i.e., cognitive impairment) that is a determinant of disability and level of community functioning. This aspect of the illness is not being addressed with current medications. Because of this and related initiatives, it is likely that individuals with schizophrenia will someday routinely receive two types of medications: one to control their psychotic symptoms, and another to improve their cognition, in order to help them benefit from non-pharmacological treatments (e.g., psychosocial rehabilitation) and more successfully reenter the community.