FDG-PET in Dementia Imaging: Whats Dragging Down Utilization?
PET imaging for Alzheimer’s disease has a fairly long history. As far back as the early 1980s, researchers realized that FDG-PET scans could show changes in brain activity indicative of Alzheimer’s disease. The Centers for Medicare and Medicaid Services (CMS), however, did not reimburse for the procedure. Consequently, when patients were referred for an imaging exam, clinicians more commonly ordered a SPECT study.
The problem with this model is that SPECT is considerably lessuseful than PET in the detection of Alzheimer’s disease, says DanSilverman, MD, PhD, head of the neuronuclear imaging centerand associate director of the imaging core of the Alzheimer’s Disease Center at University of California Los Angeles. During the 1980s and early 1990s, studies showed the accuracy of FDG-PET imaging was 15 to 20 percent greater than SPECT exams. Some patients with symptoms of Alzheimer’s who could pay for PET studies were able to access the exam if two conditions existed: the referring physician understood the benefits of PET and alocal PET provider offered brain imaging for dementia.
Fast forward to 2004. As the data demonstrating the accuracyand clinical utility of PET for Alzheimer’s disease accumulated, CMS shifted gears and approved Medicare reimbursement for theindication. Reimbursement, however, has not solved the accessproblem. In fact, PET remains heavily underutilizedin imaging patients with symptoms of Alzheimer’s disease. That’s because accessrequires both a referring physician well-acquainted with the utility of PET in the diagnosis of Alzheimer’s disease and a PET center with expertisein the use of PET for Alzheimer’s disease.
Five years after CMS approved Medicarereimbursement, Silverman believes that FDG-PET imaging could be ready for a move to the mainstream. Mostingredients for success are available. Reimbursement is relativelystable, software to facilitate image review is on the market, and the U.S. FDA is evaluating additional tracers to help physicians differentiate between Alzheimer’s disease and other neurodegenerative disorders.
Across Europe, reimbursement for FDG-PET imaging for suspected Alzheimer’s varies widely among countries—with the United Kingdom most often more restrictive, while Germany tends to be a bit more liberal, according to Karl Herholz, MD, professor of clinical neuroscience at the University of Manchester, U.K., who also has had vast clinical experience in Germany. “But across all countries,the rules are generally getting tighter,” he says.
Yet another hurdle remains in FDG-PET ’s wider proliferation in Alzheimer’s disease evaluation: the lack of effectiveness of Alzheimer’s drugs. “If you can’t treat disease, why image it?”offers Herholz. He says progress depends on more effective drugs and right now physicians across the globe are dissatisfied withthe lack of effective drugs to treat Alzheimer’s disease. Better drugs will allow a more proactive approach to treatment in the future and FDG-PET may play a role.
FDG-PET, in addition to being an effective imaging method, ispoised to play a larger role in drug development as a biomarker. Herholzcites the hope the industry holds in the Alzheimer’s Disease Neuroimaging Initiative (ADNI). The world is awaiting results from this public-private partnership sponsored by the U.S. National Institutes of Health (NIH) studying more than 800 people, half of whom have mild cognitive impairment, with the goal of determining whether brain imaging via PET and MRI , other biological markers, and clinical and neuropsychological assessment can accurately measure the progression of mild cognitive impairment and early Alzheimer’s disease. The identification of specific biomarkers of early Alzheimer’s disease and disease progression will provide a useful tool for researchers and clinicians in both the diagnosis of early Alzheimer’s disease and in the development, assessment and monitoring of new treatments. “This would be a baseline in using FD G in drug trials,” he adds. “If there is a breakthrough there, we can reassess its use in diagnostic imaging.”
Despite the clear clinical evidence showing the advantages of FDG-PET in the evaluation of patients with suspected Alzheimer’s disease, it has not yet been widely adopted. Mainstream adoption hinges on a two-pronged approach to education, says Silverman. The PET provider needs to understand the specifics of brain PET imaging. “Brain PET review is very different from oncology PET imaging,” says Silverman. In addition, the local referring physician population needs to understand the benefits of FDG-PET for patients whose symptoms suggest Alzheimer’s disease.
Silverman heads a two-day NeuroPET Preceptorship based inthe Nuclear Medicine Division at UCLA . This training programis a dedicated imaging course for radiologists, nuclear medicine physicians and technologists interested in optimizing their brain PET acquisition methods and interpretive skills. “These scans are very complicated. In oncology imaging, the tumor is metabolically active, so it shows as a hot spot. The physician isn’t just looking for hot spots on brain studies. Instead, he needs to review the activity in 40 to 50 different regions of the brain and compare results with the normal activity for each region to inform the differential diagnosis,” says Silverman.
The second day of the course focuses on software and quantification tools that can be used to assist with the evaluation. Software is used to automatically quantify activity in different regions of the brain. Other software tools analyze results on a pixel-by-pixel basis to help the reader identify regions with abnormal uptake and synaptic firing.
Software offers several advantages, says Silverman. “It can reduce the time it takes to interpret the study, draw the reader’s attention to the more subtly abnormal [but still clinically pertinent] regions of the brain that might have otherwise been overlooked and boost diagnostic confidence.”
Herholz is a large supporter of excellent and continued physiciantraining. “FDG-PET requires very well trained doctors toread complex, quantitative images of the brain,” he says. “This can be a vicious cycle [if doctor’s do not consistently read images] as they do not maintain the proper knowledge. This could affect performance. And even because this isn’t the most popular [imaging method], we need to maintain this expertise as well in the amount of manpower.” An SNM initiative is underway to focus on this.
Developing a sufficient knowledge base in the PET practice is the first step toward clinical utilization. The next, and equally important, step entails educating referring physician practicesabout brain PET . A practice can launch its brain PET program by inviting area physicians to an educational session with a brain PET expert either in a grand rounds configuration or by hosting a dinner seminar. The key, says Silverman, is to find an expert with experience as an imager who can tell a convincing story about how brain PET can help the patient and physician.
Both educational pieces, in combination with straightforward, user-friendly reports, can help establish a practice in the brain imaging realm. With a stable reimbursement situation and robust PET hardware and software, PET imaging for dementia may come into its own in the next year or two.
The problem with this model is that SPECT is considerably lessuseful than PET in the detection of Alzheimer’s disease, says DanSilverman, MD, PhD, head of the neuronuclear imaging centerand associate director of the imaging core of the Alzheimer’s Disease Center at University of California Los Angeles. During the 1980s and early 1990s, studies showed the accuracy of FDG-PET imaging was 15 to 20 percent greater than SPECT exams. Some patients with symptoms of Alzheimer’s who could pay for PET studies were able to access the exam if two conditions existed: the referring physician understood the benefits of PET and alocal PET provider offered brain imaging for dementia.
Fast forward to 2004. As the data demonstrating the accuracyand clinical utility of PET for Alzheimer’s disease accumulated, CMS shifted gears and approved Medicare reimbursement for theindication. Reimbursement, however, has not solved the accessproblem. In fact, PET remains heavily underutilizedin imaging patients with symptoms of Alzheimer’s disease. That’s because accessrequires both a referring physician well-acquainted with the utility of PET in the diagnosis of Alzheimer’s disease and a PET center with expertisein the use of PET for Alzheimer’s disease.
Five years after CMS approved Medicarereimbursement, Silverman believes that FDG-PET imaging could be ready for a move to the mainstream. Mostingredients for success are available. Reimbursement is relativelystable, software to facilitate image review is on the market, and the U.S. FDA is evaluating additional tracers to help physicians differentiate between Alzheimer’s disease and other neurodegenerative disorders.
Across Europe, reimbursement for FDG-PET imaging for suspected Alzheimer’s varies widely among countries—with the United Kingdom most often more restrictive, while Germany tends to be a bit more liberal, according to Karl Herholz, MD, professor of clinical neuroscience at the University of Manchester, U.K., who also has had vast clinical experience in Germany. “But across all countries,the rules are generally getting tighter,” he says.
Yet another hurdle remains in FDG-PET ’s wider proliferation in Alzheimer’s disease evaluation: the lack of effectiveness of Alzheimer’s drugs. “If you can’t treat disease, why image it?”offers Herholz. He says progress depends on more effective drugs and right now physicians across the globe are dissatisfied withthe lack of effective drugs to treat Alzheimer’s disease. Better drugs will allow a more proactive approach to treatment in the future and FDG-PET may play a role.
FDG-PET, in addition to being an effective imaging method, ispoised to play a larger role in drug development as a biomarker. Herholzcites the hope the industry holds in the Alzheimer’s Disease Neuroimaging Initiative (ADNI). The world is awaiting results from this public-private partnership sponsored by the U.S. National Institutes of Health (NIH) studying more than 800 people, half of whom have mild cognitive impairment, with the goal of determining whether brain imaging via PET and MRI , other biological markers, and clinical and neuropsychological assessment can accurately measure the progression of mild cognitive impairment and early Alzheimer’s disease. The identification of specific biomarkers of early Alzheimer’s disease and disease progression will provide a useful tool for researchers and clinicians in both the diagnosis of early Alzheimer’s disease and in the development, assessment and monitoring of new treatments. “This would be a baseline in using FD G in drug trials,” he adds. “If there is a breakthrough there, we can reassess its use in diagnostic imaging.”
Clinical rationale at glance
The physiological basis for imaging Alzheimer’s with FDG-PET isfairly straightforward. “FDG is biochemically similar to glucose. Unlike other organs, the brain almost exclusively uses glucose for energy. The most energy expensive thing the brain does that changes in the time frame of an imaging study is the communication of one neuron with the next—synaptic firing. Mapping the distribution of glucose or FDG in the brain provides a very good map of brain activity,” explains Silverman. Consequently, FDG-PET is an appropriate imaging study for patients whose symptoms are indicative of synaptic dysfunction. As additional therapeutic options become availablein the future, the utility of PET could be even greater. Clinical research shows that FDG-PET studies can show changes in synapticactivity years before the onset of symptoms. Early detection enables testing of new therapies and early intervention.
Understanding underutilization
Despite the clear clinical evidence showing the advantages of FDG-PET in the evaluation of patients with suspected Alzheimer’s disease, it has not yet been widely adopted. Mainstream adoption hinges on a two-pronged approach to education, says Silverman. The PET provider needs to understand the specifics of brain PET imaging. “Brain PET review is very different from oncology PET imaging,” says Silverman. In addition, the local referring physician population needs to understand the benefits of FDG-PET for patients whose symptoms suggest Alzheimer’s disease.Silverman heads a two-day NeuroPET Preceptorship based inthe Nuclear Medicine Division at UCLA . This training programis a dedicated imaging course for radiologists, nuclear medicine physicians and technologists interested in optimizing their brain PET acquisition methods and interpretive skills. “These scans are very complicated. In oncology imaging, the tumor is metabolically active, so it shows as a hot spot. The physician isn’t just looking for hot spots on brain studies. Instead, he needs to review the activity in 40 to 50 different regions of the brain and compare results with the normal activity for each region to inform the differential diagnosis,” says Silverman.
The second day of the course focuses on software and quantification tools that can be used to assist with the evaluation. Software is used to automatically quantify activity in different regions of the brain. Other software tools analyze results on a pixel-by-pixel basis to help the reader identify regions with abnormal uptake and synaptic firing.
Software offers several advantages, says Silverman. “It can reduce the time it takes to interpret the study, draw the reader’s attention to the more subtly abnormal [but still clinically pertinent] regions of the brain that might have otherwise been overlooked and boost diagnostic confidence.”
Herholz is a large supporter of excellent and continued physiciantraining. “FDG-PET requires very well trained doctors toread complex, quantitative images of the brain,” he says. “This can be a vicious cycle [if doctor’s do not consistently read images] as they do not maintain the proper knowledge. This could affect performance. And even because this isn’t the most popular [imaging method], we need to maintain this expertise as well in the amount of manpower.” An SNM initiative is underway to focus on this.
Developing a sufficient knowledge base in the PET practice is the first step toward clinical utilization. The next, and equally important, step entails educating referring physician practicesabout brain PET . A practice can launch its brain PET program by inviting area physicians to an educational session with a brain PET expert either in a grand rounds configuration or by hosting a dinner seminar. The key, says Silverman, is to find an expert with experience as an imager who can tell a convincing story about how brain PET can help the patient and physician.
Both educational pieces, in combination with straightforward, user-friendly reports, can help establish a practice in the brain imaging realm. With a stable reimbursement situation and robust PET hardware and software, PET imaging for dementia may come into its own in the next year or two.