Inadequacy of manual measurements compared to automated CT volumetry in assessment of treatment response of pulmonary metastases using RECIST criteria

Department of Radiology, Klinikum rechts der Isar, Technical University Munich, Germany.
European Radiology (Impact Factor: 4.01). 04/2006; 16(4):781-90. DOI: 10.1007/s00330-005-0036-x
Source: PubMed


The purpose of this study was to compare relative values of manual unidimensional measurements (MD) and automated volumetry (AV) for longitudinal treatment response assessment in patients with pulmonary metastases. Fifty consecutive patients with pulmonary metastases and repeat chest multidetector-row CT (median interval=2 months) were independently assessed by two radiologists for treatment response using Response Evaluation Criteria In Solid Tumours (RECIST). Statistics included relative measurement errors (RME), intra-/interobserver correlations, limits of agreement (95% LoA), and kappa. A total of 202 metastases (median volume=182.22 mm(3); range=3.16-5,195.13 mm(3)) were evaluated. RMEs were significantly higher for MD than for AV (intraobserver RME=2.34-3.73% and 0.15-0.22% for MD and AV respectively; P<0.05. Interobserver RME=3.53-3.76% and 0.22-0.29% for MD and AV respectively; P<0.05). Overall correlation was significantly better for AV than for MD (P<0.05). Intraobserver 95% LoAs were -1.85 to 1.75 mm for MD and -11.28 to 9.84 mm(3) for AV. The interobserver 95% LoA were -1.46 to 1.92 mm for MD and -11.17 to 9.33 mm(3) for AV. There was total intra-/interobserver agreement on response using AV (kappa=1). MD intra- and interobserver agreements were 0.73-0.84 and 0.77-0.80 respectively. Of the 200 MD response ratings, 28 (14/50 patients) were discordant. Agreement using MD dropped significantly from total remission to progressive disease (P<0.05). We therefore conclude that AV allows for better reproducibility of response evaluation in pulmonary metastases and should be preferred to MD in these patients.

9 Reads
  • Source
    • "However, relatively high inter- and intra-observer variability was found in the measurement of lung tumor size on CT scans, which can lead to an incorrect interpretation of tumor response [25]. Accurate and objective size measurement and detection of newly emerged metastatic nodules are of similar importance for the evaluation of therapeutic responses in malignancy [26], [27]. In clinical evaluation, objective information additional to the original CT images is highly helpful. "
    [Show abstract] [Hide abstract]
    ABSTRACT: To evaluate the accuracy of advanced non-linear registration of serial lung Computed Tomography (CT) images using Large Deformation Diffeomorphic Metric Mapping (LDDMM). FIFTEEN CASES OF LUNG CANCER WITH SERIAL LUNG CT IMAGES (INTERVAL: 62.2±26.9 days) were used. After affine transformation, three dimensional, non-linear volume registration was conducted using LDDMM with or without cascading elasticity control. Registration accuracy was evaluated by measuring the displacement of landmarks placed on vessel bifurcations for each lung segment. Subtraction images and Jacobian color maps, calculated from the transformation matrix derived from image warping, were generated, which were used to evaluate time-course changes of the tumors. The average displacement of landmarks was 0.02±0.16 mm and 0.12±0.60 mm for proximal and distal landmarks after LDDMM transformation with cascading elasticity control, which was significantly smaller than 3.11±2.47 mm and 3.99±3.05 mm, respectively, after affine transformation. Emerged or vanished nodules were visualized on subtraction images, and enlarging or shrinking nodules were displayed on Jacobian maps enabled by highly accurate registration of the nodules using LDDMM. However, some residual misalignments were observed, even with non-linear transformation when substantial changes existed between the image pairs. LDDMM provides accurate registration of serial lung CT images, and temporal subtraction images with Jacobian maps help radiologists to find changes in pulmonary nodules.
    PLoS ONE 01/2014; 9(1):e85580. DOI:10.1371/journal.pone.0085580 · 3.23 Impact Factor
  • Source
    • "Manual measurements are also subject to high inter- and intra-observer variability. Several studies have suggested that manual measurements of tumor size by radiologists are inconsistent [2], [3], [4] and should not be relied upon to provide ground truth. In response to these issues, semi-automated measurement methods have been developed to improve tumor measurement efficiency and reduce inconsistency among radiologists. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Computed tomography (CT) is a non-invasive imaging modality used to monitor human lung cancers. Typically, tumor volumes are calculated using manual or semi-automated methods that require substantial user input, and an exponential growth model is used to predict tumor growth. However, these measurement methodologies are time-consuming and can lack consistency. In addition, the availability of datasets with sequential images of the same tumor that are needed to characterize in vivo growth patterns for human lung cancers is limited due to treatment interventions and radiation exposure associated with multiple scans. In this paper, we performed micro-CT imaging of mouse lung cancers induced by overexpression of ribonucleotide reductase, a key enzyme in nucleotide biosynthesis, and developed an advanced semi-automated algorithm for efficient and accurate tumor volume measurement. Tumor volumes determined by the algorithm were first validated by comparison with results from manual methods for volume determination as well as direct physical measurements. A longitudinal study was then performed to investigate in vivo murine lung tumor growth patterns. Individual mice were imaged at least three times, with at least three weeks between scans. The tumors analyzed exhibited an exponential growth pattern, with an average doubling time of 57.08 days. The accuracy of the algorithm in the longitudinal study was also confirmed by comparing its output with manual measurements. These results suggest an exponential growth model for lung neoplasms and establish a new advanced semi-automated algorithm to measure lung tumor volume in mice that can aid efforts to improve lung cancer diagnosis and the evaluation of therapeutic responses.
    PLoS ONE 12/2013; 8(12):e83806. DOI:10.1371/journal.pone.0083806 · 3.23 Impact Factor
  • Source
    • "Analogously, more sensitive indicators of response are needed to accelerate the clinical trial process of delivering new treatments to groups of patients with unmet medical needs [6]. A potentially more sensitive and accurate alternative to line lengths as the basis for RECIST would be to measure the actual volume of the target lesions [2] [3] [7] [8]. In fact, this was proposed more than 25 years ago [9] when it was still necessary to manually demarcate the tumor boundary on each axial slice. "
    [Show abstract] [Hide abstract]
    ABSTRACT: . This study presents a semiautomated approach for volumetric analysis of lung tumors and evaluates the feasibility of using volumes as an alternative to line lengths as a basis for response evaluation criteria in solid tumors (RECIST). The overall goal for the implementation was to accurately, precisely, and efficiently enable the analyses of lesions in the lung under the guidance of an operator. Methods . An anthropomorphic phantom with embedded model masses and 71 time points in 10 clinical cases with advanced lung cancer was analyzed using a semi-automated workflow. The implementation was done using the Cognition Network Technology. Results . Analysis of the phantom showed an average accuracy of 97%. The analyses of the clinical cases showed both intra- and interreader variabilities of approximately 5% on average with an upper 95% confidence interval of 14% and 19%, respectively. Compared to line lengths, the use of volumes clearly shows enhanced sensitivity with respect to determining response to therapy. Conclusions . It is feasible to perform volumetric analysis efficiently with high accuracy and low variability, even in patients with late-stage cancer who have complex lesions.
    International Journal of Biomedical Imaging 05/2011; 2011(3):361589. DOI:10.1155/2011/361589
Show more