Groundbreaking Research Started With Skepticism

During a psychiatry elective at what was then called the Bowman Gray School of Medicine at Wake Forest University, William Carpenter visited an inpatient ward where the patients—most of whom had severe depression—were treated with electroconvulsive therapy (ECT) and sub-coma insulin.

The professor spoke convincingly, describing exactly how ECT worked.“ It was all very interesting, and I decided to do a paper on it,” Carpenter recalled. “When I finished my paper, what I'd found was that there were at least 50 different theories on how ECT worked.

“I think it was my first taste of the very big difference between what people purport to know and what they really do know.”

That same skepticism and knack for methodological precision would characterize Carpenter's inquiries for the next 40 years during a career in which he would challenge the field of psychiatry to examine what it purports to know about schizophrenia and what it really does know.

In the process he would question the notion of schizophrenia as a single, undifferentiated disease and advance the concept of separate domains of pathology, some of which—especially negative and cognitive symptoms—have proven crucial to functional outcome.

Along the way he has been a critic of the drug industry's marketing of“ me too” drugs for schizophrenia—drugs that differ little from their predecessors in terms of efficacy and are sold to a niche market of patients on the basis of side-effect profiles—and has argued for a new methodological framework for research and drug discovery.

His work has helped to move the field of schizophrenia research away from the single-disease model to one that focuses on domains of pathology within a syndrome.

One fruit of this paradigm shift has been the Measurement and Treatment Research to Improve Cognition in Schizophrenia (M ATRICS) project, a neurobiological “Manhattan Project” at NIMH bringing together industry, academic researchers, and regulatory authorities to foster development of drugs that target the cognitive impairments of schizophrenia.

“Will Carpenter is a towering figure in American psychiatry,” said past APA President Steven Sharfstein, M.D. “His contributions to the understanding of the psychopathology of the positive and negative symptoms of schizophrenia are among the most important in the last 25 years. Patients and their families are grateful for his work, researchers are grateful for his mentorship, and clinicians are grateful for his science as it has translated to practice.”

What Is the Nature of the Evidence?

The North Carolina native's first steps toward medicine and psychiatry were more fortuitous than providential—so it seemed then. A stellar receiver on his high school football team, he expected at the time to play college football and wind up a high school football coach.

But his mother took him to a vocational guidance counselor, who gave him a standard vocational aptitude test. Carpenter indicated an interest in science and music, but after the test the counselor told him he had no talent in music, was naturally lazy and would have to work hard, and should become a psychiatrist.

“We immediately set out looking for colleges with a good premed program,” Carpenter said. “I've never looked back.”

Recalling the guidance counselor's curiously specific advice, Carpenter today suspects a mother's providence. “I bet she told him to say that,” he said. “She was smart and wise and cultured and would have known what was best for me.”

Financial circumstances and love of football led him to choose a school—Wofford College in South Carolina—that offered him a football scholarship. But despite being wooed by the Baltimore Colts to go professional, he opted for medical school.

It was during his residency at the University of Rochester under the chairmanship of John Romano, M.D., that the researcher in Carpenter was kindled. And it was Romano who inspired him to study schizophrenia.

“John Romano put his personal stamp on all of us,” Carpenter told Psychiatric News. “He was a good, critical thinker who taught us to be critical of the evidence. At a time when a lot of people were practicing ideologically based therapeutics, he really emphasized asking, What is the nature of the evidence?

“He also felt strongly that schizophrenia is the disease that is central to psychiatry,” he continued. “He was very concerned that our sickest patients were in state hospitals while our best trainees would be going into private offices. He taught us that there was a moral imperative to treat the sickest.”

Following residency, Carpenter went to work at the National Institute of Mental Health (NIMH) in the depression unit under William Bunney, M.D., who is now emeritus professor of psychiatry at the University of California, Irvine. Later, he joined the International Pilot Study on Schizophrenia of the World Health Organization (WHO).

It was at WHO, as co-chief collaborating investigator in the Washington Field Research Center, that he and colleagues John Bartko, M.D., and John Strauss, M.D., published multiple papers on the signs and symptoms of schizophrenia, laying the groundwork for future research on separate domains of pathology.

In a paper in the winter 1974 Schizophrenia Bulletin,“ Speculations on the Processes That Underlie Schizophrenic Signs and Symptoms,” they posited the existence of three subdomains of disease consisting of positive symptoms, negative symptoms, and disorders of relationship. The paper was the third in a series of seminal papers on the topic.

“If... schizophrenia is a concept involving several processes, perhaps three major ones, it will be important to test the validity of these processes by providing more information about whether they have separate etiologies, outcomes, and treatment requirements,” the authors concluded then.

No Surprise About CATIE

It was also during his early training that Carpenter was exposed to a 1964 publication by John Platt in Science titled “Strong Inference,” which stated that the greatest advances in science derive from hypothesis-falsifying experiments—that is, experiments that are designed to exclude, or falsify, one or more hypotheses and then are replicated to winnow out remaining hypotheses.

It is in this way, Platt said, that researchers continually force progressive changes and refinements in theory. And it is a process, some would say, that is exactly the opposite of the way much research on schizophrenia, especially drug discovery research, has been conducted.

Fifty years of pharmaceutical-funded clinical trials tweaking existing molecules have tended to confirm the dopamine hypothesis. A theory that offers little real insight into the nature of the disease process, it rests on the observation that antipsychotic medications treat the positive psychotic symptoms of schizophrenia by acting on the dopamine system. The result has been the long line of “me too” drugs, and Carpenter has been a vocal critic of the process.

In a debate about the relative merits of first- and second-generation antipsychotics (SGAs) at APA's 2005 annual meeting, he cited a host of methodological flaws in studies purporting to show the superiority of SGAs: use of patients who had failed on first-generation drugs; the use of“ last observation carried forward” analysis to compensate for treatment drop-out and boost efficacy reporting; and the comparison of SGAs against very high doses of haloperidol (Psychiatric News, July 1, 2005).

“No one has a right to be surprised about the CATIE findings,” he told Psychiatric News more recently. “It's been easy to make drugs look good by comparing them to Haldol at the wrong dose.” (CATIE is the acronym for Clinical Antipsychotic Trials of Intervention Effectiveness.)

He added, “Almost everyone now believes the dopamine-blocking drugs treat some aspects of the disease and don't treat others. These other features are very important to functional outcome and require a different discovery pattern for creating new therapeutics that has been crystallized in the MATRICS project.”

Well Placed for Translational Research

In 1977 Carpenter became director of the Maryland Psychiatric Research Center (MPRC), one of the nation's premier psychiatric research institutions. Located on the campus of a state mental hospital, the Spring Grove Hospital in Catonsville, Md., MPRC is uniquely situated to translate basic and clinical research to the bedside.

“Being on that campus, we have a natural connection between basic science and the clinical side,” Carpenter said. “It's an intimate relationship that was designed for doing translational research even before that term was coined.”

He laments the never-ending challenge of recruiting patients for research and the chronic problem of funding. The center is a joint program of the University of Maryland and the state's Mental Hygiene Administration, with most of its budget, until last year, coming from the state.

Now, the budget allocation has shifted to the university. With just 10 percent of the center's funding in hard money, the rest comes from grants and contracts—mostly from NIMH and the National Institute on Drug Abuse.

Carpenter also worries about the increasing costs associated with patient-protection regulations.

“The costs of doing business keep going up along with the amount of time and effort required to know how to deal effectively with requirements of review boards,” Carpenter said. “There is an increasing demand for demonstrating compliance with ever-expanding requirements that are often put in place without regard to benefit-risk analysis or evidence that they are advancing the safety of research subjects.”

But MPRC is also well placed to pursue the lines of research that Carpenter believes are the future of schizophrenia research—lines that he helped to establish over a 40-year career: hypothesis-falsifying experiments aimed at drug discovery for separate domains of pathology, especially cognitive and negative symptoms.

With a scientific consensus forming around the signs and symptoms of a prodromal phase of schizophrenia, Carpenter foresees the possibility of therapeutic interventions long before the first acute psychosis.

“If people put the resources into developing drugs for negative symptoms, cognition, and other unique features of the disease, it is bound to pay off,” he said. “Suppose between ages 5 and 10 a child with a genetic predisposition for schizophrenia loses five IQ points. If we had safe, effective drugs for cognition, we would be justified in intervening early in life.

“Imagine if we could keep that child from losing those five points. It becomes a lot easier to imagine a robust effect on long-term outcome by intervening for those traits that appear years before psychosis, regardless of whether or not it actually prevents psychosis.” ▪