F-18 플로르베타벤 뇌 양전자단층촬영

Abstract

평가배경 F-18 플로르베타벤 뇌 양전자단층촬영(F-18 Florebetaben (FBB) brain positron emission tomography (PET)) 검사는 인지장애(의심) 환자를 대상으로 F-18 플로르베타벤을 주입한 후 뇌 양전자방출 단층촬영 또는 양전자방출전산화단층촬영을 시행하여 베타아밀로이드 신경반 밀도를 측정하고 평가함으로써 알츠하이머 질병 및 기타 원인의 인지장애를 감별 및 판별하기 위한 검사이다. F-18 플로르베타벤 뇌 PET 검사는 2015년 제3차 신의료기술평가위원회(2015.3.27.)에서 안전성 및 효과성이 있는 기술로 심의되어 보건복지부 고시(제2015-73호, 2015.5.11.) 후 비급여로 등재되었다(2016.7.1.). 보건의료 자원의 효율적 사용도모와 의료기술의 최적 사용을 유도하기 위해 최신 근거를 평가할 필요가 있는 안건으로 제안되었으며, 2021년 제10차 의료기술재평가위원회(2021.10.15.)에서 의료기술재평가 안건으로 선정되어 재평가를 수행하였다.

평가방법 F-18 플로르베타벤 뇌 PET 검사의 안전성 및 효과성 평가를 위해 체계적 문헌고찰을 수행하였다. 모든 평가방법은 평가 목적을 고려하여 “F-18 플로르베타벤/F-18 플루트메타몰 뇌 양전자단층촬영 소위원회(이하 ‘소위원회’라 한다)”의 논의를 거쳐 확정하였다. 소위원회는 핵의학과 2인, 정신건강의학과 2인, 신경과 2인, 영상의학과 1인, 근거기반의학 1인 등 총 8인으로 구성하였다. 평가의 핵심질문은 “인지장애(의심) 환자에서 알츠하이머 질병 및 기타 원인의 인지장애 진단 및 예측목적의 F-18 플로르베타벤 뇌 양전자단층촬영 검사가 임상적으로 안전하고 효과적인가?”이었고, 안전성은 검사 및 F-18 플로르베타벤 약물 관련 이상반응을 결과지표로, 효과성은 진단정확성, 진단변화, 예측정확성, 치료변화와 비용 등을 결과지표로 평가하였다. 체계적 문헌고찰은 핵심질문을 토대로 국외 3개, 국내 5개 데이터베이스에서 검색하여 문헌선정 및 배제기준에 따라 두 명의 검토자가 독립적으로 선별하고 선택하였다. 문헌의 비뚤림위험 평가는 Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2)를 사용하였고, 자료추출은 문헌의 연구국가, 대상자 특성, 안전성 지표, 효과성 지표 등을 사전에 정한 양식에 추출하였다. 비뚤림위험 평가와 자료추출은 두 명의 검토자가 독립적으로 수행하였으며, 의견 불일치한 경우 논의하여 합의하였다. 자료분석은 정량적 분석(quantitative analysis)이 가능한 경우 통합정확도를 제시하였고, 정량적 분석이 불가능한 경우 정성적(qualitative review) 분석과 숲 그림(forest plot)을 제시하였다. 본 평가는 소위원회의 결론을 검토하여 의료기술재평가위원회에서 최종심의 후 권고등급을 제시하였다.

평가결과 F-18 플로르베타벤 뇌 PET 검사의 최종 선택 문헌은 14편이었고, 안전성과 효과성 등 임상적 결과지표를 보고한 문헌은 13편(12개 연구)이었고, 비용-효과 결과를 보고한 문헌은 1편이었다. 대부분 문헌에서 F-18 플로르베타벤 뇌 PET 검사는 참고표준인 임상 진단과 비교해 세부 진단 감별 및 인지장애 질환의 진단정확도를 확인하는 보조검사로 활용되었고, 대체가능성을 검토하기 위해 F-18 플로르베타벤 뇌 PET 검사와 F-18 플루데옥시글루코스(Fluorodeoxyglucose, FDG) 뇌 PET 검사를 비교한 문헌은 1편이었다. 비뚤림위험 평가결과, ‘환자 선택’과 관련된 비뚤림위험 및 적용가능성에 대한 우려가 높은 연구는 주로 환자-대조군 연구였고, 그 외 ‘중재검사’, ‘참고표준검사’, ‘연구진행과 시점’ 등은 비뚤림위험이 낮았다.

안전성 F-18 플로르베타벤 뇌 PET 검사의 안전성에 대해 보고한 문헌은 14편 중 3편이었고, 3편 모두 심각한 부작용은 발생하지 않았고, 일부에서 약물 투여경로와 관련된 주사 부위 이상반응이 발생하였으나 방사선 관련 과민반응과는 관련이 없었고, 대부분의 부작용은 경증에서 중등도였고, 연구기간동안 회복되었다.

효과성 F-18 플로르베타벤 뇌 PET 검사의 알츠하이머병 치매(Alzheimer's dementia, AD) 관련 진단정확도는 6편에서 보고하였고, 민감도 0.67~0.97, 특이도 0.45~1.00, ROC 곡선의 아래 면적(Area under the receiver operating characteristic (ROC) curve, AUC)은 0.62~0.90이었다. 통합민감도 0.82 (95% 신뢰구간(Confidence interval, CI) 0.73~0.88), 통합특이도 0.89 (95% CI 0.69~0.96), 통합AUC 0.89 (95% CI 0.85~0.91)였다. 알츠하이머병 치매가 아닌 기타 치매, 전두측두엽 치매 등의 진단정확도는 3편에서 보고하였고, 민감도 0.15~1.00, 특이도 0.71~1.00, AUC 0.31~0.90이었고, 치매 종류의 구분없이 전체 치매에 대한 진단정확도 결과(3편), 민감도 0.67~0.78, 특이도 0.50~1.00, AUC 0.63~0.75였다. F-18 플로르베타벤 뇌 PET 검사의 기억성 경증인지장애(Amnestic mild cognitive impairment, aMCI)의 진단 결과를 보고한 문헌은 2편이었고, 민감도 0.65~0.72, 특이도 0.58, AUC 0.63이었다. MCI 진단정확도는 3편의 문헌에서 보고하였고, 민감도 0.60~0.84, 특이도 0.72~0.84, AUC 0.68~0.81이었다. 주관적 인지감퇴(Subjective cognitive decline, SCD) 진단에 대해 보고한 문헌은 1편으로 민감도 0.38, 특이도 0.79, AUC 0.75였다. F-18 플로르베타벤 뇌 PET 검사 후 진단변화에 대해 보고한 문헌은 3편이었고, AD 환자에서 11.1%, 전두측두엽 치매 환자에서 28.6%, 기타 치매 환자에서 10.0%, MCI 환자에서 18.8%가 진단이 변경되었다. F18 플로르베타벤 뇌 PET 검사의 예측정확도는 3편에서 보고하였고, 1년 치매 예측정확도는 민감도 0.86, 특이도 0.78, AUC 0.81이었고, 2년 치매 예측정확도는 민감도 0.90, 특이도 0.76, AUC 0.82였고, 3년 치매 예측정확도는 민감도 0.50, 특이도 0.85, AUC 0.68이었고, 4년 치매 예측정확도는 민감도 0.81, 특이도 0.84, AUC 0.82였다. F-18 플로르베타벤 뇌 PET 검사 후 치매환자의 30.3%, 66.8%에서 치료변화가 있었고(2편), MCI 환자의 29.9%에서 치료변화, SCD 환자의 10.7%에서 치료변화가 있었다. 약물복용은 치매 환자의 21.4%, MCI 환자의 22.2%, SCD 환자의 2.5%에서 추가되었다. 경제성 평가 문헌에서는 F-18 플로르베타벤 뇌 PET 검사를 기존 치매 진단과정에 추가했을 때 비용-효과를 분석하였고, 기본분석 결과, 치매 전 코호트는 1인당 총 $13,018 절감, 치매 코호트는 총 $11,389 비용절감을 기대할 수 있는 것으로 보고하였다. 민감도 분석에서는 F-18 플로르베타벤 뇌 PET 검사가 우세 전략이 되기 위해서는 치매 확진까지의 시간을 40% 이상 앞당겨야 한다고 보고하였다.

결론 및 제언 F-18 플로르베타벤 뇌 PET 검사 소위원회는 재평가 결과에 근거하여 다음과 같이 결론을 도출하였다. F-18 플로르베타벤 뇌 PET 검사는 일부 환자에서 약물 투입경로 관련 이상반응이 발생하였으나 증상이 경미하였고, 모두 회복되었다. 효과성 지표인 진단정확도는 알츠하이머병 치매 관련 진단정확도는 0.62~0.90, 통합AUC 0.89, 비알츠하이머병 치매 진단정확도는 0.31~0.90, 경증인지장애는 0.68~0.81로 알츠하이머병 치매의 진단정확도가 가장 높았다. F-18 플로르베타벤 뇌 PET 검사를 통해 알츠하이머병 치매 환자의 11%, 경증인지장애 환자의 18%에서 진단이 변경되었다. 1년 치매 예측정확도 0.81, 2년 치매 예측정확도 0.82, 3년 치매 예측정확도 0.68, 4년 치매 예측정확도 0.82였다. 또한, F-18 플로르베타벤 뇌 PET 검사 결과가 양성인 경우 치료계획에 반영되거나 약물 복용이 증가하였다. 소위원회는 이를 종합하여 F-18 플로르베타벤 뇌 PET 검사는 인지장애 및 치매 진단에서 보조검사로 주로 활용되며, 비교적 안전한 검사로 평가하였다. 알츠하이머병 치매 진단시 문헌간 이질성이 일부 존재했지만, 통합민감도 0.82, 통합특이도 0.89, 통합 AUC 0.89로 높은 진단정확도를 보이고, 치료계획 및 추적관찰 등을 고려할 때 효과적인 검사로 평가하였다. 소위원회는 알츠하이머병 외 다른 종류의 치매 및 인지장애 질환에서는 일부 연구들에서 높은 진단정확도를 보고하였으나, 결과가 일관되지 않아 개별 질환별로 결론을 내리기에는 임상근거가 부족하였다고 평가하였다. 또한, 소위원회는 F-18 플로르베타벤 뇌 PET 및 F-18 플루트메타몰 뇌 PET 검사의 최적 사용을 위해서는 구체적인 적용 대상과 임상 적용 시기, 의료이용 경로, 비용-효과성 등에 대한 추가연구가 필요하다고 제언하였다. 2022년 제5차 의료기술재평가위원회(2022.5.13.)에서는 소위원회 검토 결과를 논의하여 의료기술재평가사업 관리지침 제4조 제10항에 의거 “F-18 플로르베타벤 뇌 양전자단층촬영”에 대해 다음과 같이 심의하였다. 의료기술재평가위원회는 F-18 플로르베타벤 뇌 PET 검사는 안전하고 알츠하이머병 치매 및 의심 환자에서 진단 및 예측의 보조검사로 효과적이라고 심의하였고, 알츠하이머병외 치매와 인지장애에 서는 근거가 불확실하여 결론 도출에 다양한 이견이 있음을 확인하였다. 또한, 검사비용이 100만원 이상의 고가인 점과 진단의 보조적 검사로 활용되고, 오ㆍ남용의 우려 등에 대한 의견이 있었다. 이러한 의견을 바탕으로 의료기술재평가위원회는 F-18 플로르베타벤 뇌 양전자단층촬영 검사는 알츠하이머병 치매와 의심환자에서 ‘조건부 권고함’으로 심의하였다.

주요어 F-18 플로르베타벤 뇌 양전자단층촬영 검사, 인지장애, 알츠하이머병 치매, 안전성, 진단정확도

Background The “F-18 Florbetaben (FBB) Brain Positron Emission Tomography (PET)” scan is an examination for differentiated detection and differential diagnosis between Alzheimer's Disease and cognitive impairment of other causes by injecting F-18 florbetaben and conducting brain PET or positron emission computed tomography to measure and assess the β-amyloid neuritic plaque density for patients (suspects) with cognitive impairment. F-18 FBB brain PET was approved as a safe and effective technology at the 3rd New Health Technology Assessment Committee (March 27, 2015) in 2015 and notified by the Ministry of Health and Welfare (Notice No. 2015-73, May 11, 2015), followed by listing as a non-benefit item for national health insurance coverage (July 1 2016). In order to promote efficient utilization of healthcare resources and induce optimal use of health technology, this technology was proposed as an item that required assessment of the latest evidence. This technology was selected to be subject to health technology reassessment (HTR) at the 10th Health Technology Reassessment Committee in 2021 (October 15, 2021) and HTR for the technology was performed.

Methods A systematic review was conducted for assessment of safety and effectiveness of F-18 florbetaben (FBB) brain PET. All assessment methods were finalized following the review by the F-18 FBB/F-18 Flutemetamol Brain PET Sub-committee (Hereinafter Sub-committee) in consideration of the purpose of the assessment. The Sub-committee consisted of a total of eight members: two in Nuclear Medicine, two in Neuropsychiatry, two in Neurology, one in Radiology, and one in Evidence-Based Medicine. The key question for HRT was: “Is F-18 FBB brain PET scan clinically safe and effective for diagnosis and prediction of Alzheimer’s disease and cognitive impairment of other causes in patients with cognitive impairment (suspects)?” Assessment was conducted on the safety of the technology with adverse events related to the examination and adverse reactions to F-18 FBB as outcome measures and the effectiveness of the technology was assessed with diagnostic accuracy, changes in diagnosis, prediction accuracy, changes in treatments, and cost as outcome measures. In the process of systematic review, based on the above key question, three international and five domestic databases were used for literature search and the articles were selected based on independent screening by two reviewers according to the inclusion/exclusion criteria for the related literature. A tool of Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) was used for risk of bias assessment of the literature. For data extraction, information on countries where the research was carried out, patient characteristics, outcome measures for safety and effectiveness, etc. were extracted using a predefined format. Risk of bias assessment and data extraction were conducted independently by two reviewers, and in case of disagreement, the two reviewers held discussions to draw consensus on the matter. For data analysis, pooled accuracy was presented when quantitative analysis was possible, and qualitative review and forest plots were presented otherwise. In this assessment, the conclusion drawn by the Sub-committee was reviewed, and upon final review at the Health Technology Reassessment Committee, the grade of recommendation was presented.

Results The number of articles on F-18 FBB brain PET scan finally selected through the above process was 14, the number of articles that reported clinical outcome measures such as safety and effectiveness was 13 (12 studies), and there was one article that reported outcome measures related to cost-effectiveness. In most of the literature, F-18 FBB brain PET was used as an adjunct technique for confirming diagnostic accuracy related to cognitive impairment and for detailed differential diagnosis in comparison with clinical diagnosis, the reference standard. There was one article that compared F-18 FBB brain PET with F-18 fluorodeoxyglucose (FDG) brain PET to examine the possibility of substitution of the technique. As a result of risk of bias assessment, studies with high risk of bias related to patient selection were mainly case-control studies, and index test, reference standard, study progress and timepoint showed a low risk of bias.

Safety Out of 14 articles, three publications reported on the safety of F-18 FBB brain PET scan, and no serious events were reported in all of the three studies. In some cases, injection site reactions related to the route of administration occurred but these were not related to radiation-induced hypersensitivity reactions, and most of the adverse events were mild to moderate in severity and resolved during the study period.

Effectiveness The diagnostic accuracy of F-18 FBB brain PET scan related to Alzheimer's dementia (AD) was reported in six articles, with sensitivity in the range of 0.67 - 0.97, specificity of 0.45 to 1.00, and area under the receiver operating characteristic (ROC) curve (AUC) was 0.62 to 0.90. The pooled sensitivity was 0.82 (95% Confidence Interval (CI) of 0.73 to 0.88), pooled specificity was 0.89 (95% CI in the range of 0.69 - 0.96), and pooled AUC was 0.89 (95% CI in the range of 0.85- 0.91). The diagnostic accuracy of dementia other than AD such as frontotemporal dementia was reported in three articles, with sensitivity 0.15 to 1.00, specificity 0.71 to 1.00, and AUC 0.31 to 0.90. The diagnostic accuracy for all-cause dementia (regardless of the type of dementia) was reported in three articles, with sensitivity 0.67to0.78, specificity 0.50to1.00, and AUC 0.63to0.75. There were two articles that reported on diagnosis of amnestic mild cognitive impairment (aMCI) using F-18 FBB brain PET, and sensitivity was in the range of 0.65-0.72 with specificity of 0.58 and AUC of 0.63. Three articles reported on diagnostic accuracy of mild cognitive impairment (MCI), with sensitivity in the range of 0.60 to 0.84, specificity 0.72 to 0.84, and AUC 0.68 to 0.81. One article reported on diagnosis of subjective cognitive decline (SCD), with sensitivity 0.38, specificity 0.79, and AUC 0.75. There were three articles that reported changes in diagnosis after the F-18 FBB brain PET imaging, and the diagnosis was changed in 11.1% of patients with AD, 28.6% of patients with frontotemporal dementia, 10.0% for other types of dementia, and 18.8% for patients with MCI. The prediction accuracy of the examination using F-18 FBB brain PET was reported in three articles. The prediction accuracy of 1-year dementia incidence showed sensitivity 0.86, specificity 0.78, and AUC 0.81; the prediction accuracy of 2-year dementia incidence showed sensitivity 0.90, specificity 0.76, and AUC 0.82; the prediction accuracy of 3-year dementia incidence showed sensitivity 0.50, specificity 0.85, and AUC 0.68; and the prediction accuracy of 4-year dementia incidence showed sensitivity 0.81, specificity 0.84, and AUC 0.82. After the examination using F-18 FBB brain PET, two articles reported that there were changes in treatments for 30.3% and 66.8% of patients with dementia, respectively. The ratio of changes in treatments was 29.9% in patients with MCI and 10.7% in patients with SCD. Additional medication was prescribed in 21.4% of patients with AD, 22.2% of MCI patients, and 2.5% of patients with SCD. In the literature that evaluated cost effectiveness of the technology, cost effects were analyzed when the F-18 FBB brain PET was added to the conventional procedure of dementia diagnosis. Results from basic analysis reported that the introduction of the technology would contribute to the cost saving by $13,018 per person for pre-dementia cohort, and $11, 389 for dementia cohort. In the sensitivity analysis, it was reported that the time to confirmatory diagnosis of dementia should be shortened by more than 40% in order for the F-18 FBB brain PET imaging to become a dominant strategy for diagnosis of dementia.

Conclusion and Suggestions The F-18 FBB Brain PET Sub-committee derived the following conclusions based on the results of reassessment of the technology. The use of F-18 FBB brain PET involved occurrence of adverse events related to drug administration route, but the symptoms were mild and the AEs were resolved over time. As for the diagnostic accuracy, one of the outcome measures for effectiveness, the diagnostic accuracy related to AD was in the range of 0.62 to 0.90 with pooled AUC 0.89. The diagnostic accuracy related to dementia not caused by Alzheimer’s disease was in the range of 0.31 to 0.90, and that for MCI was 0.68 to 0.81, indicating the diagnostic accuracy was highest for patients with AD. Through the F-18 FBB brain PET imaging, 11% of the AD patients and 18% of the MCI patients underwent changes in their diagnosis. The prediction accuracy of 1-year dementia incidence was 0.81, and the accuracy was 0.82 for 2-year dementia incidence, 0.68 for 3-year dementia incidence, and 0.82 for 4-year dementia incidence. In addition, when the result of F-18 FBB brain PET imaging was positive, it was incorporated into the treatment plan or the prescription of medication to be administered increased accordingly. In comprehensive consideration of the above results, the assessment by the Sub-committee concluded that F-18 FBB brain PET method is mainly used as an adjunct technique for diagnosis of cognitive impairment and dementia, and that the technology is relatively safe to use. In diagnosis of AD, although there were some discrepancies depending on the literature of previous studies, high diagnostic accuracy was reported with pooled sensitivity 0.82, pooled specificity 0.89, and pooled AUC 0.89; thus the technology was assessed as an effective technique when considering treatment plans and follow-up monitoring/observations for patients. Also, the Sub-committee assessed that although some of the existing works reported high diagnostic accuracy for types of dementia other than AD and cognitive impatient, the results were not consistent across these previous studies; thus, it was determined that the technology requires further clinical evidence to draw conclusions for each type of dementia/cognitive impairment. In addition, the Sub-committee suggested that for optimal application of F-18 FBB brain PET and F-18 flutemetamol brain PET, further studies are needed on specific patient groups for application of the technology, timing of clinical application, route of using medical services, cost-effectiveness, etc. Based on the results of the review by the Sub-committee at the 6th Health Technology Reassessment Committee (June 10, 2022) in 2022, the following is the result of the review for “F-18 florbetaben brain PET” in accordance with Article 4, Paragraph 10 of the Management Guidelines for Health Technology Reassessment Project. The review by the Health Technology Reassessment Committee determined that F-18 FBB brain PET is a safe technology and is effective as an adjunct technique for diagnosis and prediction for patients with AD and adults with suspected AD. However, for types of dementia other than AD and cognitive impairment, the evidence for the technology was uncertain, leading to different opinions in deriving a conclusion for these cases. In addition, there were opinions that expressed concerns about the high cost of examination in its use as an adjunct method for diagnosis, which is more than one million KRW, and also concerns about likelihood of misuse/overuse of the technology. Based on these opinions, the review by the Health Technology Reassessment Committee determined the grade of recommendation for F-18 FBB brain PET as “conditional recommendation” for patients with AD and adults with suspected AD.

Keywords: F-18 Florbetaben (FBB) Brain positron emission tomography, Cognitive impairment, Alzheimer's dementia, Dementia, Safety, Diagnostic accuracy