DOI: 10.1148/rg.282075068
RadioGraphics 2008;28:329-344
© RSNA, 2008
Radiologic Measurements of Tumor Response to Treatment: Practical Approaches and Limitations1
Chikako Suzuki, MD,
Hans Jacobsson, MD, PhD,
Thomas Hatschek, MD, PhD,
Michael R. Torkzad, MD, PhD,
Katarina Bodén, MD,
Yvonne Eriksson-Alm, RT,
Elisabeth Berg, BSc,
Hirofumi Fujii, MD, PhD,
Atsushi Kubo, MD, PhD, and
Lennart Blomqvist, MD, PhD
1 From the Department of Diagnostic Radiology, Institution for Molecular Medicine and Surgery, Karolinska University Hospital Solna and Karolinska Institute, Stockholm S-171 76, Sweden (C.S., H.J., M.R.T., K.B., Y.E.A., L.B.); the Department of Oncology, Karolinska University Hospital Solna (T.H.); the Medical Statistics Unit, Department of Learning, Informatics, Management and Ethics, Karolinska Institute (E.B.); the Department of Diagnostic Radiology, National Cancer Center East, Chiba, Japan (H.F.); and the Department of Radiology, Keio University School of Medicine, Tokyo, Japan (A.K.). Presented as an education exhibit at the 2006 RSNA Annual Meeting. Received April 9, 2007; revision requested May 15 and received July 12; accepted July 18. Supported by a Grant-in-Aid for Center of Excellence Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan. All authors have no financial relationships to disclose.
Address correspondence to C.S. (e-mail: Chikako.Tanaka{at}ki.se).
 |
Abstract
|
|---|
Objective response assessment is important to describe the treatment effect of anticancer drugs. Standardization by using a "common language" is also important for comparison of results from different trials. In contrast to clinical results, which can be subjective, diagnostic imaging provides a greater opportunity for objectivity and standardization. It was generally accepted that a decrease in tumor size correlated with treatment effect; as a result, imaging was adopted for lesion measurement in the World Health Organization (WHO) criteria in 1979. However, because of some limitations of the WHO criteria, the Response Evaluation Criteria in Solid Tumors (RECIST) were introduced in 2000. In RECIST, imaging was recognized as indispensable for response evaluation of solid tumors. Nevertheless, the widespread use of multidetector computed tomography and other imaging innovations have made RECIST outdated, with a concomitant need for modifications. Meanwhile, newer anticancer agents with targeted mechanisms of action have demonstrated an inherent limitation and unsuitability of anatomic tumor evaluation that assesses only lesion size. In addition, the effect of these new drugs changes the paradigm according to which tumor response or response rate is measured. Complete and partial responses cannot be the end points in all clinical trials; in some cases, disease control or progression-free survival may be the more relevant end point.
© RSNA, 2008
|
Introduction
|
|---|
Highly consistent, reproducible, objective, and standardized response criteria as a "common language" are indispensable to evaluate the effects of new therapies in multicenter trials. This is especially true when new treatments are compared to previously established therapies. Moreover, standardized criteria would prove fundamental when two therapies are given longitudinally (ie, consecutively) and there is a need to isolate the effect of each treatment for better comparison. Development of new drugs is expensive, time-consuming, and laborious (1,2); by determining if the drug is effective or ineffective, better use could be made of resources.
Response evaluation with diagnostic imaging has evolved over the past 25 years. Initially, the World Health Organization (WHO) response criteria were introduced in 1979 without any specific imaging stipulations or protocols (3). Therefore, different groups subsequently proposed modifications that may have led to confusion (4), with the criteria not being applied consistently across all studies.
To unify and standardize the criteria, the Response Evaluation Criteria in Solid Tumors (RECIST) were introduced in 2000 (5) by a task force set up by the European Organization for Research and Treatment of Cancer (EORTC), the National Cancer Institute of the United States, and the National Cancer Institute of Canada. RECIST is useful because the response assessment criteria are standardized as a common language. The pivotal role of imaging in response assessment is recognized, and specific imaging guidelines are defined. However, practical use of RECIST together with the rapid development of imaging techniques and pharmaceuticals have highlighted the limitations of RECIST and the need for updated criteria. For this reason, accurate knowledge of current imaging standards in relation to inherent limitations of RECIST is needed.
In this article, we review the process of how to evaluate tumor response with RECIST and also with the corresponding WHO criteria, especially from the radiologic point of view. Practical approaches and limitations of RECIST and WHO criteria as well as future perspectives on tumor response evaluation are also discussed.
From the WHO Criteria to RECIST
In 1979, uniform criteria were proposed that standardized the recording and reporting of response, recurrence, and disease-free interval and the grading of acute and subacute toxicity in solid tumor treatment (3). These criteria (WHO criteria) were based on bi- or two-dimensional (2D) measurements because it was not possible to measure tumor volume with the imaging technology available at that time (Fig 1). These criteria were based on the assumption that the tumor is spherical and has a circular cross section (Table 1) (6,7). These criteria have received wide acceptance and have become known as the WHO criteria for reporting the results of cancer treatment.
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Methods of tumor measurement according to the RECIST and WHO criteria. With the WHO criteria, the longest diameter (A) and the longest perpendicular diameter (B) are obtained and multiplied (2D measurement). With RECIST, only the longest diameter (A) is obtained (uni- or one-dimensional [1D] measurement). The location of the longest diameter is decided independently of previous study results when the tumor changes shape or rotates.
|
|
However, WHO criteria do not mention the minimum lesion size or the number of lesions to be selected in patients with multiple lesions. Nor do the WHO criteria consider the type of imaging modality that should be used. Progressive disease (PD), originally defined as a 25% increase in the product of 2D diameters, was defined by some investigators as the increase in the sum of all lesions and by others as the increase in any one lesion (Fig 2) (4). Measurement error in a single lesion, especially when it is small, could heavily impact a decision on progression. Patients might be overestimated as having PD and inappropriately removed from a beneficial treatment regimen. Application of the WHO criteria includes sources of variability and potential for overestimation (8).
View larger version (51K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Illustrative chart for a practical approach to evaluating tumors according to RECIST and WHO criteria. Data are from a patient with multiple lung, liver, lymph node, and bone metastases. Each number under "Site*" corresponds to the above-mentioned number of an organ; for example, (2) = lymph node and (4) = liver. Values in parentheses under "Size" (#) represent the longest perpendicular diameter of the target lesion for 2D (WHO) evaluation. Baseline evaluation: A, Note the presence or absence of tumor involvement. B, Start from the largest lesion. C, Note in what window setting the lesion is measured (eg, SW = soft-tissue window). Follow-up evaluation: D, Before starting measurement, look for new lesions or unequivocal progression of nontarget lesions that are critical to follow-up evaluation. The patient will then automatically be classified as having PD regardless of other measurements. E, Even if the target lesions become too small to obtain accurate measurements, it is important to continue reporting them. F, In the case of disease progression, the smallest sum of longest diameters since the trial started should be referenced rather than the baseline sum of the longest diameters. G, Because lymph node metastases often change size along their short axis, measurement of the longest diameter in the axial plane may neglect a change in lymph node size. In the WHO criteria, if the product of the longest diameter and longest perpendicular diameter increases more than 25% in any lesion, the patient should be categorized as having PD. On the other hand, in RECIST, progression in any lesion without an overall 20% increase in the sum of longest diameters is not considered progression. In the case shown, the lymph node metastasis increased more than 25% at the third follow-up. According to the WHO criteria, the patient should be classified as having PD. However, the sum of the longest diameters decreased more than 30% from the baseline sum; according to RECIST, the patient is classified as having a partial response (PR). Ao = aortic, N = no, Rt = right, S = segment, SD = stable disease, U = unknown, Y = yes, yy-mm-dd = year-month-day.
|
|
In 2000, the RECIST guidelines were introduced by the EORTC, the U.S. National Cancer Institute, and others (5). The primary goal of RECIST was to try to unify the various modifications of the WHO criteria so that meaningful comparisons could be made among studies. This included the following: (a) the need to maintain the four categories of responses (complete response [CR], partial response [PR], stable disease [SD], and PD); (b) the need to maintain the same definition of PR so that favorable results of future therapies can be compared with WHO criteria even though the measurements will be different (Table 1); and (c) the need to modify the definition of PD.
There are five major differences between RECIST and the WHO criteria: (a) 1D measurements are adopted, which encourages measurement of more lesions and minimizes labor (6); (b) the type of imaging to be used is stipulated; (c) the types of tumors that should or should not be chosen are defined; (d) the number of tumor lesions used for assessment is specified; and (e) the cutoff point for definition of PD is larger.
As shown in Table 2, the cutoff for PD in RECIST is a 20% increase in the sum of the longest diameters. According to WHO criteria, a 25% increase is considered PD. Note that an increase in one diameter (RECIST) corresponds to a 73% increase in spherical volume, whereas a 25% increase in 2D measurements (WHO) corresponds to a 40% increase in spherical volume (Table 1) (6). Furthermore, instead of the increase in a single lesion being considered, in RECIST the sum of the longest diameters of a limited number of target lesions is used. Despite these differences, various studies have demonstrated good concordance between WHO criteria and RECIST for overall response (5,6,9,10). However, discrepancies have been demonstrated for time to progression; RECIST requires a larger increase in lesions and a longer delay to detect disease progression (Fig 2) (4,11,12).
View this table:
[in this window]
[in a new window]
|
Table 2. Comparison of Evaluation Methods and Definitions of Lesion and Response Categories in the RECIST and WHO Criteria
|
|
The full texts of the RECIST criteria as well as more detailed explanations corresponding to many practical questions are available on the EORTC RECIST Web site (http://www.eortc.be/Recist/Default.htm) (13). The RECIST quick reference is also available on the National Cancer Institute Web site (http://imaging.cancer.gov/clinicaltrials/imaging) (14).
Table 2 provides a summary of the RECIST and WHO criteria, and Table 3 presents the definition of overall response in RECIST. Figure 2 provides an example of an evaluation procedure according to the RECIST and WHO criteria.
View this table:
[in this window]
[in a new window]
|
Table 3. Overall Response for All Possible Combinations of Tumor Responses in Target and Nontarget Lesions with or without the Appearance of New Lesions
|
|
|
The Process: How to Evaluate Tumor Response with RECIST
|
|---|
Measurable Lesions
It is necessary to first evaluate if existing tumor lesions are measurable or not. A measurable lesion size is defined as a longest diameter of 20 mm or greater at nonspiral CT with 10-mm section thickness or at chest radiography and a longest diameter of 10 mm or greater at spiral CT with 5-mm reconstructed section thickness in the axial plane (not the sagittal or coronal planes at CT or MR imaging). RECIST requires that the minimum size of target lesions should be no less than double the section thickness to minimize "partial volume" effects.
RECIST implies that the imaging modality should be a highly reproducible and objective one such as CT or MR imaging, although there is no detailed specification on how to use these modalities. On the other hand, US, mammography, endoscopy, and laparoscopy are no longer accepted because of their subjective nature and low reproducibility. CT is the imaging modality most frequently used for this purpose. However, the recent advancements in multidetector CT and MR imaging raise new issues, demanding a revision of RECIST (15). In general, CT scans today are performed and reconstructed with thinner sections and higher spatial resolution than RECIST originally required. It would be unacceptable to have inferior imaging resolution simply to adhere to the literal definition of RECIST. However, the concept of standardizing minimum lesion size and section thickness is important to be able to make longitudinal comparisons in large multicenter trials.
Nonmeasurable Lesions
Small lesions (<10 mm), skeletal metastases, leptomeningeal disease, ascites, pleural or pericardial effusion, inflammatory breast disease, lymphangitis cutis or pulmonis, abdominal masses that are not confirmed and followed up with imaging techniques, cystic or necrotic lesions, and tumor lesions situated in a previously irradiated area are all considered to be nonmeasurable.
Target Lesions
Measurable lesions, up to a maximum of five lesions per organ and 10 lesions total, representative of all involved organs, should be selected and identified as target lesions. They should be selected on the basis of size (those with the longest diameter) and suitability for accurate repeated measurements. In this sense, lesions in mobile organs (eg, gastrointestinal tract, ovaries) might be inappropriate (Fig 3). It is obviously necessary to identify findings that may mimic tumors, such as a focal spared area in a fatty liver or a pseudo liver lesion around the round ligament.
View larger version (161K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Inappropriate selection of lesions in a 56-year-old patient with breast cancer and bilateral ovarian metastases. (a) Axial contrast-enhanced CT scan from the baseline study shows enlargement and heterogeneous enhancement of both ovaries (arrowheads). (b) Corresponding CT scan obtained after therapy shows the left ovary (arrow) turned behind the uterus. Evaluation is thus more difficult due to organ movement. Arrowhead = right ovary.
|
|
View larger version (160K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Inappropriate selection of lesions in a 56-year-old patient with breast cancer and bilateral ovarian metastases. (a) Axial contrast-enhanced CT scan from the baseline study shows enlargement and heterogeneous enhancement of both ovaries (arrowheads). (b) Corresponding CT scan obtained after therapy shows the left ovary (arrow) turned behind the uterus. Evaluation is thus more difficult due to organ movement. Arrowhead = right ovary.
|
|
To avoid generating inconsistencies in the evaluation, it is best to choose a well-defined, isolated lesion. However, metastases are often ill-defined and confluent (Fig 4), which may provide different definitions of tumor margins and the longest diameter. Second, the selection of target lesions in patients with multiple metastases is a source of variability itself.
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Difficulty in measuring the tumor. (a) Axial contrast-enhanced CT scan of a 58-year-old patient with colon cancer shows a lobulated and ill-defined liver metastasis. The lesion seems to be the result of fusion of two metastases as well. (b) Axial contrast-enhanced CT scan of a 74-year-old patient with rectal cancer shows confluent liver metastases. Different numbers of lesions, although with a maximum of five, and different combinations of lesions can be selected for baseline measurement and follow-up according to RECIST. Furthermore, RECIST does not specify how a confluent lesion should be measured if it separates into several lesions after treatment.
|
|
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Difficulty in measuring the tumor. (a) Axial contrast-enhanced CT scan of a 58-year-old patient with colon cancer shows a lobulated and ill-defined liver metastasis. The lesion seems to be the result of fusion of two metastases as well. (b) Axial contrast-enhanced CT scan of a 74-year-old patient with rectal cancer shows confluent liver metastases. Different numbers of lesions, although with a maximum of five, and different combinations of lesions can be selected for baseline measurement and follow-up according to RECIST. Furthermore, RECIST does not specify how a confluent lesion should be measured if it separates into several lesions after treatment.
|
|
RECIST limits the number of target lesions without scientific evidence or theoretical justification (16). On the other hand, the minimum number of tumors is also unspecified. Although Zacharia et al (17) have shown that a limited number of lesions, for instance, one or two lesions, might give the same result in response evaluation, there remains some doubt whether a small number of lesions appropriately represent a patient’s real response. To measure all tumors would be ideal because lesions may respond differently, but this would be impractical and in some cases impossible. The larger the number of lesions that are measured, the fewer false evaluations are generated (8,18). Mazumdar et al (16) provide a theoretical (mathematical) approach that can define the appropriate number of lesions that should be selected in patients with multiple lesions. However, even with this approach, the number of lesions being measured needs more consideration and discussion.
Nontarget Lesions
Nontarget lesions include both measurable and nonmeasurable lesions. Measurable lesions that exceed the maximum acceptable number of target lesions (ie, = five in the same organ or = 10 in the body) are thus included in the group referred to as nontarget lesions. Nontarget lesions do not need to be measured in follow-up studies, but any change should be noted. One should clearly record the presence or absence of tumor involvement (Fig 2). Final response categorization should be accomplished by evaluating changes in both target and nontarget lesions as well as noting the presence or absence of new lesions (Table 3).
Considerations in Regard to Specific Organs
Lymph Node and Adrenal Gland Metastases.—
In lymph node metastases, in contrast to metastases to other organs, the metastatic lesion itself is usually not visualized at imaging. The enlargement of a lymph node is used as a surrogate indicator of metastasis. If size is used as an indicator of metastasis, the short axis should be considered, since this is the best predictor of the presence of metastatic disease (19–23). As shown in Figure 5, a lymph node frequently grows or shrinks along its short axis. According to RECIST, a lymph node can be included as a target lesion. However, measuring the longest diameter of lymph node metastases in the axial plane according to RECIST involves the risk of neglecting the described pattern of response in lymph nodes.
View larger version (122K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Measurement of a lymph node metastasis in a 58-year-old patient with breast cancer, a lymph node metastasis, and multiple liver metastases. (a) Axial contrast-enhanced CT scan shows a lymph node in the portocaval region (arrow) and multiple ill-defined liver metastases. (b) Corresponding CT scan obtained after therapy shows that the lymph node has decreased in size along its short axis and become flattened (arrow) without a change in its longest diameter. RECIST requires only 1D measurement (ie, the longest diameter in the axial plane); thus, elliptical changes in lesions during treatment are not considered.
|
|
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Measurement of a lymph node metastasis in a 58-year-old patient with breast cancer, a lymph node metastasis, and multiple liver metastases. (a) Axial contrast-enhanced CT scan shows a lymph node in the portocaval region (arrow) and multiple ill-defined liver metastases. (b) Corresponding CT scan obtained after therapy shows that the lymph node has decreased in size along its short axis and become flattened (arrow) without a change in its longest diameter. RECIST requires only 1D measurement (ie, the longest diameter in the axial plane); thus, elliptical changes in lesions during treatment are not considered.
|
|
Adrenal metastases are observed in approximately 27% of patients with malignant tumors at autopsy (Fig 6) (24,25). On the other hand, the prevalence of an unexpected adrenal nodule or mass (incidentaloma) has been reported to be 0.35%–5.0% at CT among all patients referred for CT (26), and the prevalence of metastases among these patients is in the range of 38%–57% (27–29). Recent advances in imaging techniques help in differentiation of metastases from incidentally detected adenomas by demonstrating the presence of fat in the latter (25).
View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. Appearance of a new lesion in the adrenal gland in a 49-year-old man with rectal cancer. (a, b) Axial contrast-enhanced CT scans (lung window) obtained before (a) and after (b) treatment show multiple lung metastases. The target lesion (arrowhead) and many other lesions have decreased in size after treatment. (c, d) Axial contrast-enhanced CT scans of the abdomen, obtained before (c) and after (d) treatment, show a new lesion in the adrenal gland after treatment (arrow in d). This finding should be interpreted as a case of PD.
|
|
View larger version (133K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. Appearance of a new lesion in the adrenal gland in a 49-year-old man with rectal cancer. (a, b) Axial contrast-enhanced CT scans (lung window) obtained before (a) and after (b) treatment show multiple lung metastases. The target lesion (arrowhead) and many other lesions have decreased in size after treatment. (c, d) Axial contrast-enhanced CT scans of the abdomen, obtained before (c) and after (d) treatment, show a new lesion in the adrenal gland after treatment (arrow in d). This finding should be interpreted as a case of PD.
|
|
View larger version (130K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6c. Appearance of a new lesion in the adrenal gland in a 49-year-old man with rectal cancer. (a, b) Axial contrast-enhanced CT scans (lung window) obtained before (a) and after (b) treatment show multiple lung metastases. The target lesion (arrowhead) and many other lesions have decreased in size after treatment. (c, d) Axial contrast-enhanced CT scans of the abdomen, obtained before (c) and after (d) treatment, show a new lesion in the adrenal gland after treatment (arrow in d). This finding should be interpreted as a case of PD.
|
|
View larger version (129K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6d. Appearance of a new lesion in the adrenal gland in a 49-year-old man with rectal cancer. (a, b) Axial contrast-enhanced CT scans (lung window) obtained before (a) and after (b) treatment show multiple lung metastases. The target lesion (arrowhead) and many other lesions have decreased in size after treatment. (c, d) Axial contrast-enhanced CT scans of the abdomen, obtained before (c) and after (d) treatment, show a new lesion in the adrenal gland after treatment (arrow in d). This finding should be interpreted as a case of PD.
|
|
RECIST does not specify which imaging criteria should be used to define lymph nodes or adrenal glands that contain metastases and when biopsy should be performed. Furthermore, when lymph nodes and adrenal glands remain visible, their mere presence indicates that the treatment result could not be categorized as CR.
Bone Metastases.—
Bone metastasis is the only metastatic site in approximately 20%–30% of patients with breast or prostate cancer (30). If RECIST is applied to categorize treatment response, these patients would not be enrolled in experimental studies that require measurable lesions according to RECIST. This is because a bone metastasis is defined as a nonmeasurable lesion (5). It is quite common that bone metastases change from osteolytic to sclerotic or from sclerotic to osteolytic without changes in size. At CT, differentiation between active metastatic disease and responding bone metastases may therefore be difficult. Responding metastases may become more sclerotic without change in size. Recent studies have shown the promise of a combination of different parameters, not only size, provided by different imaging modalities (MR imaging, skeletal scintigraphy, and positron emission tomography [PET]) for assessing the treatment response of bone metastases (31,32). Therefore, this issue is important to consider when a revision of RECIST or new criteria are introduced.
Cystic or Necrotic Metastases.—
Many metastatic lesions contain cystic or necrotic parts. According to RECIST, cystic lesions are considered nonmeasurable (5,15). This does not mean that one should refrain from assessing lesions that are not primarily solid, yet contain foci of necrosis or cystic components. RECIST does not specifically deal with how to assess these lesions. In addition, RECIST does not take into account changes in attenuation at CT or signal intensity at MR imaging when lesions are followed up during or after treatment.
Measurement of Target Lesions in the Baseline Study
Measurement of only the longest diameter of target lesions in the axial plane is required for RECIST (Fig 1). The longest diameter is used as a substitute for tumor volume because it was not practically possible to measure tumor volume with imaging at the time RECIST was introduced. However, widespread use of multidetector CT, MR imaging, and post–image processing procedures enables the radiologist to view lesions from any arbitrary plane, to measure the "true" longest diameter of a lesion, and even to accurately measure the volume (three-dimensional [3D] measurement) (33,34).
The volumetric approach is attractive and may be the most reliable method to reflect real changes in size (33). Volumetric change of the tumor during treatment has been shown to correlate with patient prognosis (35–37). Some studies have shown good correlation of 3D volume measurements with 1D and 2D measurements (38–40). On the other hand, other studies have shown little or no agreement between 1D, 2D, and 3D measurements and little or no benefit of 3D measurement in predicting prognosis (12,38). However, the latter data were obtained before the advent of multidetector CT, and further studies might be warranted. This is discussed later in the section on future perspectives.
|
Treatment Evaluation: Importance of a Consistent Imaging Protocol throughout the Study
|
|---|
Imaging Protocol in Regard to Contrast Enhancement
In RECIST, enhancement with intravenous contrast material is stipulated as the preferred method, but the vascular phases that should be used are not specified. Carcinoid tumors, breast cancer, and gastrointestinal stromal tumors (GISTs) tend to generate hypervascular liver metastases that are best imaged during the arterial phase. Other tumors, such as colorectal cancer, generate liver metastases that are mainly hypovascular. The metastases may both be erroneously measured and escape detection if they are evaluated in a suboptimal phase of contrast enhancement. To make matters worse, after treatment, the optimal phase of enhancement for metastases may shift during follow-up (Fig 7). Some authors recommend imaging during three enhancement phases at baseline and whenever possible at follow-up in cases of GISTs (41). However, this recommendation may not apply to all cases and may increase radiologic work and radiation exposure. A disease-specific trial approach has to be considered.
View larger version (146K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. Issues in the timing of contrast enhancement in a 64-year-old woman with rectal cancer. (a) Axial contrast-enhanced CT scan from the baseline study shows liver metastases (arrow). Because metastases from colorectal cancer are mainly hypovascular, they are well demonstrated during the portal venous or equilibrium phase (a, c) but not during the arterial phase (b). (b) On an arterial phase CT scan obtained 8 weeks after treatment, the indicated lesion (arrow) seems to have shrunk and other lesions seem to have disappeared. (c) On a portal venous phase CT scan obtained 12 weeks after treatment, the indicated lesion (arrow) demonstrates an increase in size relative to that in the previous study; this finding should not be interpreted as representing disease progression. In this case, treatment-related fatty infiltration has also reduced the contrast between the lesion and the liver parenchyma, thus making it more difficult to define the borders of the tumor. Inconsistency in the enhancement phase used during follow-up may preclude accurate comparison.
|
|
View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. Issues in the timing of contrast enhancement in a 64-year-old woman with rectal cancer. (a) Axial contrast-enhanced CT scan from the baseline study shows liver metastases (arrow). Because metastases from colorectal cancer are mainly hypovascular, they are well demonstrated during the portal venous or equilibrium phase (a, c) but not during the arterial phase (b). (b) On an arterial phase CT scan obtained 8 weeks after treatment, the indicated lesion (arrow) seems to have shrunk and other lesions seem to have disappeared. (c) On a portal venous phase CT scan obtained 12 weeks after treatment, the indicated lesion (arrow) demonstrates an increase in size relative to that in the previous study; this finding should not be interpreted as representing disease progression. In this case, treatment-related fatty infiltration has also reduced the contrast between the lesion and the liver parenchyma, thus making it more difficult to define the borders of the tumor. Inconsistency in the enhancement phase used during follow-up may preclude accurate comparison.
|
|
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7c. Issues in the timing of contrast enhancement in a 64-year-old woman with rectal cancer. (a) Axial contrast-enhanced CT scan from the baseline study shows liver metastases (arrow). Because metastases from colorectal cancer are mainly hypovascular, they are well demonstrated during the portal venous or equilibrium phase (a, c) but not during the arterial phase (b). (b) On an arterial phase CT scan obtained 8 weeks after treatment, the indicated lesion (arrow) seems to have shrunk and other lesions seem to have disappeared. (c) On a portal venous phase CT scan obtained 12 weeks after treatment, the indicated lesion (arrow) demonstrates an increase in size relative to that in the previous study; this finding should not be interpreted as representing disease progression. In this case, treatment-related fatty infiltration has also reduced the contrast between the lesion and the liver parenchyma, thus making it more difficult to define the borders of the tumor. Inconsistency in the enhancement phase used during follow-up may preclude accurate comparison.
|
|
Other Imaging Parameters
It is not only vascularity that may cause differences when measuring tumors. Many studies have demonstrated that the imaging acquisition and display protocols have a large impact on tumor measurement, and different settings may cause about a 50% difference in size of the same tumor (42–44). In a recent study, significant differences in volume measurements according to tumor sizes, ill- or well-defined margins, and CT section thickness were demonstrated; of interest, volume may even differ depending on the software used for measurements (45).
In clinical trial settings, standardization of imaging is mandatory to allow comparison of results. Workshops and continuous education for radiologists can facilitate the management of imaging within clinical trials. An interactive Web-based reporting and monitoring system might also be a solution for education and quality control. One could also propose to assign a dedicated radiologist at each site within a multicenter trial to be able to facilitate follow-up of changes over time to reduce variability in results due to multiple observers. This strategy is practiced in our department.
Measurement of Target Lesions during Follow-up Evaluations
Lesions should be measured at the same window setting in each examination. However, all standard window settings should be used for general evaluation. For example, mediastinal lymphadenopathy might be missed inadvertently with the lung window and bone metastases would often be undetectable with the soft-tissue window.
The longest diameter in the axial plane is always measured no matter how the longest axis differs at follow-up. Confluent lesions pose a problem in follow-up studies. There is no consensus about how to measure whether a confluent lesion became separate lesions or whether separate lesions became confluent. According to the RECIST questions and answers (13), the longest diameter of split lesions should be measured, summed up, and reported as one lesion. However, it is uncertain whether these methods can reflect the tumor response.
Ill-defined lesions, surrounded by a ground-glass opacity area, or spiculations, remain difficult to measure (46–48). In fact, these findings are frequently observed in most metastases, and they change their appearance subsequent to therapy. It is also frequently seen that a measurable lesion shrinks and become ill-defined and difficult to appropriately measure after treatment (Fig 8). As stated on the RECIST questions and answers Web site, some authors insist on continuing measurement, even if lesions become very small or measure zero (13). In such cases, inconsistencies in measurements due to partial volume effects may not be taken into account.
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8a. Measurement of an ill-defined lung metastasis in a 64-year-old woman with breast cancer. (a) Axial CT scan (lung window) from the baseline study shows a well-defined metastasis in the peripheral part of the left lower lung (arrow). (b) Axial CT scan obtained after three cycles of treatment shows that the lesion (arrow) has decreased in size. However, because the borders of the lesion have become ill-defined and spiculated, it is difficult to define the margin and the longest diameter. (c) CT scan obtained after five cycles of treatment shows that the lesion has developed into an area of ground-glass opacity (arrow). No imaging modality allows determination of whether an area of ground-glass opacity contains tumor cells, and there is no consensus on how to measure or describe such lesions.
|
|
View larger version (122K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8b. Measurement of an ill-defined lung metastasis in a 64-year-old woman with breast cancer. (a) Axial CT scan (lung window) from the baseline study shows a well-defined metastasis in the peripheral part of the left lower lung (arrow). (b) Axial CT scan obtained after three cycles of treatment shows that the lesion (arrow) has decreased in size. However, because the borders of the lesion have become ill-defined and spiculated, it is difficult to define the margin and the longest diameter. (c) CT scan obtained after five cycles of treatment shows that the lesion has developed into an area of ground-glass opacity (arrow). No imaging modality allows determination of whether an area of ground-glass opacity contains tumor cells, and there is no consensus on how to measure or describe such lesions.
|
|
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8c. Measurement of an ill-defined lung metastasis in a 64-year-old woman with breast cancer. (a) Axial CT scan (lung window) from the baseline study shows a well-defined metastasis in the peripheral part of the left lower lung (arrow). (b) Axial CT scan obtained after three cycles of treatment shows that the lesion (arrow) has decreased in size. However, because the borders of the lesion have become ill-defined and spiculated, it is difficult to define the margin and the longest diameter. (c) CT scan obtained after five cycles of treatment shows that the lesion has developed into an area of ground-glass opacity (arrow). No imaging modality allows determination of whether an area of ground-glass opacity contains tumor cells, and there is no consensus on how to measure or describe such lesions.
|
|
What has been discussed in this section are some examples of problems and solution presented on the EORTC RECIST Web site (http://www.eortc.be/Recist/Default.htm) (13). The reader is urged to consult this site for other questions.
Frequency of Follow-up Studies
RECIST requires follow-up of every other cycle of chemotherapy (ie, 6–8 weeks). An additional examination at least 4 or more weeks after demonstrated CR or PR is required to confirm the response. It seems that after the introduction of RECIST, radiologic examinations, most commonly CT, have increased in frequency. At our center, we try to have a dedicated radiologist incorporated already in the initial phases of study planning to optimize the radiologic protocols throughout the trial. The ALARA (as low as reasonably achievable) principle should always be applied (49), and radiologists must take the lead in promoting this principle.
|
Categorization of Overall Response
|
|---|
Appearance of New Lesions or Unequivocal Progression of Nontarget Lesions
The appearance of a new lesion (Fig 6) or progression of nontarget lesions (Fig 9) has a great impact on response evaluation independent of the method of selecting or measuring a lesion. Once a new lesion appears, the tumor response is classified as PD. In RECIST, there is no specific limit on the number or minimum size of new lesions (13). It is important to focus on a search for new lesions and for progression of nontarget lesions. Therasse et al (50) demonstrated that 58% of cases of PD were due to the appearance of new lesions.
View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9a. Unequivocal progression of nontarget lesions in a 45-year-old woman with breast cancer. Axial contrast-enhanced CT scans obtained before (a) and after (b) therapy show progression of nontarget lesions. The descriptor "unequivocal progression of nontarget lesions" should be strictly restricted to cases with obvious progression that requires no measurement.
|
|
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9b. Unequivocal progression of nontarget lesions in a 45-year-old woman with breast cancer. Axial contrast-enhanced CT scans obtained before (a) and after (b) therapy show progression of nontarget lesions. The descriptor "unequivocal progression of nontarget lesions" should be strictly restricted to cases with obvious progression that requires no measurement.
|
|
However, these findings should be clearly identified and strictly limited to cases without any doubt, because once response is categorized as PD, the ongoing treatment will be disrupted and potentially stopped. In certain cases, treatment-related changes in existing lesions may be misinterpreted. For example, although a bone metastasis is one of the nonmeasurable lesions, an osteoblastic reaction due to therapy after the development of osteolytic lesions represents a favorable result, but may be falsely interpreted as PD (Fig 10). It is also well known that liver metastases from GISTs treated with imatinib mesylate (Gleevec; Novartis Pharmaceuticals, East Hanover, NJ) show decreased attenuation, which demarcates these lesions more strongly in contrast to the normal liver tissue. These reactions can mimic new lesions (Fig 11) and may be falsely evaluated as PD.
View larger version (90K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10a. Osteoblastic changes after therapy in a 62-year-old patient with metastatic rectal cancer. (a) Axial CT scan (bone window) shows bilateral osteolytic lesions in the iliac bones (arrows). (b) Corresponding CT scan obtained after therapy shows larger osteoblastic lesions with periosteal thickening in the left iliac bone (arrowhead). Osteoblastic lesions stand out or become larger relative to osteolytic lesions and might be mistakenly evaluated as new lesions or progression of nontarget lesions, causing the case to be classified as PD. However, osteoblastic reactions after therapy in breast or colon cancer are frequently a treatment effect.
|
|
View larger version (90K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10b. Osteoblastic changes after therapy in a 62-year-old patient with metastatic rectal cancer. (a) Axial CT scan (bone window) shows bilateral osteolytic lesions in the iliac bones (arrows). (b) Corresponding CT scan obtained after therapy shows larger osteoblastic lesions with periosteal thickening in the left iliac bone (arrowhead). Osteoblastic lesions stand out or become larger relative to osteolytic lesions and might be mistakenly evaluated as new lesions or progression of nontarget lesions, causing the case to be classified as PD. However, osteoblastic reactions after therapy in breast or colon cancer are frequently a treatment effect.
|
|
View larger version (149K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11a. Liver metastases in a 72-year-old man with GIST. The metastases were treated with imatinib mesylate. (a) Axial contrast-enhanced CT scan obtained before treatment show multiple liver metastases. (b) CT scan obtained after treatment shows that tiny existing lesions (arrows) are highlighted because of treatment effect, resulting in better contrast between the lesions and normal liver tissue. This finding should be strictly distinguished from the appearance of new lesions, which automatically causes the case to be categorized as PD. Furthermore, treatment effect does not always produce shrinkage of lesions. Changes in tumor attenuation rather than tumor size are more relevant in this case.
|
|
View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11b. Liver metastases in a 72-year-old man with GIST. The metastases were treated with imatinib mesylate. (a) Axial contrast-enhanced CT scan obtained before treatment show multiple liver metastases. (b) CT scan obtained after treatment shows that tiny existing lesions (arrows) are highlighted because of treatment effect, resulting in better contrast between the lesions and normal liver tissue. This finding should be strictly distinguished from the appearance of new lesions, which automatically causes the case to be categorized as PD. Furthermore, treatment effect does not always produce shrinkage of lesions. Changes in tumor attenuation rather than tumor size are more relevant in this case.
|
|
Limitations of Basic Concepts of Tumor Measurement
Reduction in tumor size does not always represent tumor response. The proposition that a decrease in tumor size corresponds to an improvement in the prognosis is not true for all cases. Exceptions necessitate a reevaluation, and new criteria have to be introduced (51).
Sometimes size is inversely correlated with treatment effect. In liver metastases from testicular tumors as well as from GISTs, it is well known that metastases may become larger after treatment due to cystic degeneration or hemorrhage occurring within the tumor (52). In liver metastases from GISTs, a nodule-within-a-mass pattern without a corresponding change in tumor size is the most important indication of progression (53); in these metastases, treatment effect correlates with attenuation (in Hounsfield units) rather than tumor size (Fig 11) (41,52). Criteria recently developed by Choi (41), which combine tumor size and attenuation for evaluation of GISTs, demonstrated a correlation between response and the time to progression. Furthermore, the combined criteria allowed more accurate prediction of overall survival than did RECIST. Some forms of treatment may lead to treatment response without a concomitant decrease in tumor size. This would lead these cases to be classified as stable disease according to RECIST or the WHO criteria; hence, one might erroneously underestimate treatment response. In these cases, the overall disease control rate, the progression-free survival, or time to progression would be more valuable than the objective response rate (CR = PR).
In previously reported meta-analyses, good correlation between tumor response according to WHO criteria or RECIST and patients’ outcomes has been demonstrated (54,55). However, in recent studies that evaluated targeted drugs, little or no correlation between tumor response and survival has been shown (56–58). In a trial that evaluated use of erlotinib in non–small cell lung cancer, the observed response rate was less than 10% despite a significant prolongation of overall survival in the erlotinib group (58).
Current evaluation criteria based on morphologic evaluation have limited ability for evaluation of targeted drugs (58,59). More integrated evaluation criteria based on morphologic changes will be needed in the future, including not only tumor size but also results of functional imaging. A new surrogate indicator of response will also be needed.
|
Future Perspectives
|
|---|
Multidetector CT and MR Imaging for Tumor Evaluation
Adhering strictly to RECIST might preclude us from taking advantage of the most recent advancements in multidetector CT and MR imaging. The RECIST criterion of 10 mm as the minimum lesion size with spiral CT might not allow use of thinner sections or evaluation in other planes than the true axial. Use of thinner sections makes it possible to use smaller lesions at baseline for future follow-up. In addition, with thinner sections, we may be able to continue measuring lesions even after they become smaller than 10 mm during and after treatment. Despite the introduction of 64-plus row multidetector CT, there is most probably going to be a minimal lesion size that we must adhere to even in the future.
In addition, RECIST considers only the size of lesions. With advancements in MR imaging and functional imaging and the introduction of new agents and new sequences that allow evaluation of aspects other than size, other features might have an important role in response evaluation.
Integration of PET and PET/CT for Tumor Evaluation
There is no doubt that functional evaluation such as PET will provide a novel surrogate end point for assessing the clinical efficacy of new agents that target defined tumor biologic reactions. Although PET is costly, availability is limited, and sensitivity for detecting changes in lesions smaller than 1 cm is uncertain (60), many studies have shown that fluorine 18 (18F) fluorodeoxyglucose (FDG) PET is of value in evaluating not only targeted drugs for GIST metastases but other drugs as well (61,62). Furthermore, the efficacy of FDG PET has also shown promise for monitoring treatment response for earlier detection of responders. This may increase the possibility of shortening trial times (63,64). Early identification of responders and nonresponders is important in order to select patients who might not experience any therapeutic benefit, to consider alternative therapies, and to avoid toxic effects; early identification will also minimize the expense of clinical trials.
As yet, no international evaluation guidelines exist for functional imaging in the assessment of response to treatment in solid tumors. Unless widely accepted guidelines are introduced, PET will not gain acceptance as a tool for functional evaluation of therapy. EORTC and U.S. National Cancer Institute guidelines for PET stipulate and recommend how FDG PET should be performed and the results translated for evaluation (65,66). In both guidelines, the standardized uptake value is recognized as valuable in the evaluation of FDG uptake in PET images. However, the standardized uptake value is not an independent value but is dependent on uptake time, the patient’s body weight (which often changes during therapy), blood glucose levels, definition of regions of interest, attenuation corrections, and the effects of metabolism from other normal tissues. Furthermore, the wide spread of PET/CT necessitates new considerations. For acquisition of functional information only, the use of CT in PET/CT can be limited to attenuation correction, which entails use of low-dose CT without contrast enhancement. If diagnostic CT is required, then one may perform CT with contrast enhancement as part of PET/CT examination (64).
The issues that need to be addressed are as follows: (a) definition of cutoff points for defining treatment-related response, (b) time periods for evaluation (67), and (c) optimal utilization of both PET and CT components of PET/CT. Monitoring tumor response with PET will be indispensable in the future but is still in its infancy.
General Considerations
RECIST was intended originally to minimize work overload by adopting 1D measurements and limiting the maximum number of lesions; however, this was unsuccessful. RECIST requires specialized knowledge and labor for the radiologist. Also, evaluation of lesions that were erroneously selected or measured may contribute to a false impression of objectivity in an inappropriate categorization of treatment effect. There is also a risk that erroneously evaluated lesions at baseline may be falsely evaluated again during follow-up if the radiologist is biased by the baseline evaluation.
Many studies have focused only on the appropriate methodology of tumor evaluation: how to measure lesions and how many lesions should be measured. Aside from the methodology, it is indispensable to consider what is most relevant to the patient’s benefit. It is important to recognize that measuring tumor size is not always equivalent to evaluating tumor response. Recent imaging developments allow us to perform correct, precise, and reproducible measurement based not only on tumor morphology but also on tumor function. For this reason, new criteria for tumor response assessment will need to integrate measurements with morphologic and functional characteristics that can be standardized (68). The focus should be on how to measure tumor response rather than simply on how to measure tumors. Whatever the new critera will be in the future is matter for debate, but one thing is definitive: this is going to be a stepwise ever-changing project adapting to new techniques and new treatments.
Finally, the reader is cautioned that RECIST, WHO, and any other criteria developed are intended for clinical trials. Despite this, both radiologists and clinicians use or ask for these measurements in daily practice. Although RECIST may offer some summarized view of the total response, its use should not replace complete radiologic evaluation for an individual patient.
|
Conclusions
|
|---|
The increasing demand for objectively verifying the treatment response of new anticancer agents has resulted in an increased use of medical imaging. Despite the limitations of RECIST, there is a high potential level of standardized assessment of the response to treatment of solid tumors. However, advances in medical technology are on the point of revolutionizing the response evaluation. The practical approaches presented and limitations with the current criteria must be considered in a revision of RECIST and future criteria.
|
Acknowledgments
|
|---|
The authors thank Boel Söderen, MD, Veli Söderlund, MD, PhD, and Staffan Bremmer, MD, PhD, for their assistance with image preparation and all their colleagues in the Department of Diagnostic Radiology, Karolinska University Hospital Solna. The authors also thank Barbara L. Clough for her outstanding linguistic review.
|
Footnotes
|
|---|
Abbreviations: CR = complete response, EORTC = European Organization for Research and Treatment of Cancer, FDG = fluorine 18 fluorode-oxyglucose, GIST = gastrointestinal stromal tumor, 1D = one-dimensional, PD = progressive disease, PR = partial response, RECIST = Response Evaluation Criteria in Solid Tumors, SD = stable disease, 3D = three-dimensional, 2D = two-dimensional, WHO = World Health Organization
|
References
|
|---|
- DiMasi JA, Hansen RW, Grabowski HG. The price of innovation: new estimates of drug development costs. J Health Econ 2003;22(2):151–185.[CrossRef][Medline]
- Seddon BM, Workman P. The role of functional and molecular imaging in cancer drug discovery and development. Br J Radiol 2003;76(spec no 2):S128–S138.[Abstract/Free Full Text]
- World Health Organization. WHO handbook for reporting results of cancer treatment. http://whqlibdoc.who.int/publications/9241700483.pdf.
- Mazumdar M, Smith A, Schwartz LH. A statistical simulation study finds discordance between WHO criteria and RECIST guideline. J Clin Epidemiol 2004;57(4):358–365.[CrossRef][Medline]
- Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92(3):205–216.[Abstract/Free Full Text]
- James K, Eisenhauer E, Christian M, et al. Measuring response in solid tumors: unidimensional versus bidimensional measurement. J Natl Cancer Inst 1999;91(6):523–528.[Abstract/Free Full Text]
- Padhani AR, Ollivier L. The RECIST (Response Evaluation Criteria in Solid Tumors) criteria: implications for diagnostic radiologists. Br J Radiol 2001;74(887):983–986.[Free Full Text]
- Warr D, McKinney S, Tannock I. Influence of measurement error on assessment of response to anticancer chemotherapy: proposal for new criteria of tumor response. J Clin Oncol 1984;2(9):1040–1046.[Abstract]
- James K, Eisenhauer E, Therasse P. Re: Measure once or twice: does it really matter? J Natl Cancer Inst 1999;91(20):1780–1781.[Free Full Text]
- Prasad SR, Jhaveri KS, Saini S, Hahn PF, Halpern EF, Sumner JE. CT tumor measurement for therapeutic response assessment: comparison of unidimensional, bidimensional, and volumetric techniques—initial observations. Radiology 2002; 225(2):416–419.[Abstract/Free Full Text]
- Tran LN, Brown MS, Goldin JG, et al. Comparison of treatment response classifications between unidimensional, bidimensional, and volumetric measurements of metastatic lung lesions on chest computed tomography. Acad Radiol 2004;11(12): 1355–1360.[CrossRef][Medline]
- Warren KE, Patronas N, Aikin AA, Albert PS, Balis FM. Comparison of one-, two-, and three-dimensional measurements of childhood brain tumors. J Natl Cancer Inst 2001;93(18):1401–1405.[Abstract/Free Full Text]
- European Organization for Research and Treatment of Cancer. EORTC RECIST Web site, questions and answers. http://www.eortc.be/Recist/Default.htm.
- National Cancer Institute of the United States. RECIST quick reference. http://imaging.cancer.gov/clinicaltrials/imaging.
- Husband JE, Schwartz LH, Spencer J, et al. Evaluation of the response to treatment of solid tumours: a consensus statement of the International Cancer Imaging Society. Br J Cancer 2004; 90(12):2256–2260.[Medline]
- Mazumdar M, Smith A, Debroy PP, Schwartz LH. A theoretical approach to choosing the minimum number of multiple tumors required for assessing treatment response. J Clin Epidemiol 2005;58(2):150–153.[CrossRef][Medline]
- Zacharia TT, Saini S, Halpern EF, Sumner JE. CT of colon cancer metastases to the liver using modified RECIST criteria: determining the ideal number of target lesions to measure. AJR Am J Roentgenol 2006;186(4):1067–1070.[Abstract/Free Full Text]
- Schwartz LH, Mazumdar M, Brown W, Smith A, Panicek DM. Variability in response assessment in solid tumors: effect of number of lesions chosen for measurement. Clin Cancer Res 2003;9(12): 4318–4323.[Abstract/Free Full Text]
- Cheson BD, Horning SJ, Coiffier B, et al. Report of an international workshop to standardize response criteria for non-Hodgkin’s lymphomas. NCI Sponsored International Working Group. J Clin Oncol 1999;17(4):1244. [Published correction appears in J Clin Oncol 2000;18(11):2351.][Abstract/Free Full Text]
- Dorfman RE, Alpern MB, Gross BH, Sandler MA. Upper abdominal lymph nodes: criteria for normal size determined with CT. Radiology 1991; 180(2):319–322.[Abstract/Free Full Text]
- Einstein DM, Singer AA, Chilcote WA, Desai RK. Abdominal lymphadenopathy: spectrum of CT findings. RadioGraphics 1991;11(3):457–472.[Abstract]
- Kim SH, Kim SC, Choi BI, Han MC. Uterine cervical carcinoma: evaluation of pelvic lymph node metastasis with MR imaging. Radiology 1994;190(3):807–811.[Abstract/Free Full Text]
- van den Brekel MW, Castelijns JA, Snow GB. Detection of lymph node metastases in the neck: radiologic criteria. Radiology 1994;192(3):617–618.[Free Full Text]
- Chong S, Lee KS, Kim HY, et al. Integrated PET-CT for the characterization of adrenal gland lesions in cancer patients: diagnostic efficacy and interpretation pitfalls. RadioGraphics 2006;26(6): 1811–1824; discussion 1824–1826.[Abstract/Free Full Text]
- Elsayes KM, Mukundan G, Narra VR, et al. Adrenal masses: MR imaging features with pathologic correlation. RadioGraphics 2004;24(suppl 1): S73–S86.[Abstract/Free Full Text]
- Cook DM, Loriaux DL. The incidental adrenal mass. Am J Med 1996;101(1):88–94.[CrossRef][Medline]
- Bernardino ME, Walther MM, Phillips VM, et al. CT-guided adrenal biopsy: accuracy, safety, and indications. AJR Am J Roentgenol 1985;144(1): 67–69.[Abstract/Free Full Text]
- Silverman SG, Mueller PR, Pinkney LP, Koenker RM, Seltzer SE. Predictive value of image-guided adrenal biopsy: analysis of results of 101 biopsies. Radiology 1993;187(3):715–718.[Abstract/Free Full Text]
- Welch TJ, Sheedy PF 2nd, Stephens DH, Johnson CM, Swensen SJ. Percutaneous adrenal biopsy: review of a 10-year experience. Radiology 1994; 193(2):341–344.[Abstract/Free Full Text]
- Therasse P, Eisenhauer EA, Verweij J. RECIST revisited: a review of validation studies on tumour assessment. Eur J Cancer 2006;42(8):1031–1039.[CrossRef][Medline]
- Hamaoka T, Madewell JE, Podoloff DA, Hortobagyi GN, Ueno NT. Bone imaging in metastatic breast cancer. J Clin Oncol 2004;22(14):2942–2953.[Abstract/Free Full Text]
- Tombal B, Rezazadeh A, Therasse P, Van Cangh PJ, Vande Berg B, Lecouvet FE. Magnetic resonance imaging of the axial skeleton enables objective measurement of tumor response on prostate cancer bone metastases. Prostate 2005;65(2):178–187.[CrossRef][Medline]
- Bornemann L, Kuhnigk JM, Dicken V, et al. New tools for computer assistance in thoracic CT. II. Therapy monitoring of pulmonary metastases. RadioGraphics 2005;25(3):841–848.[Abstract/Free Full Text]
- Kuhnigk JM, Dicken V, Bornemann L, et al. Morphological segmentation and partial volume analysis for volumetry of solid pulmonary lesions in thoracic CT scans. IEEE Trans Med Imaging 2006; 25(4):417–434.[CrossRef][Medline]
- Andreopoulou E, Andreopoulos D, Adamidis K, et al. Tumor volumetry as predictive and prognostic factor in the management of ovarian cancer. Anticancer Res 2002;22(3):1903–1908.[Medline]
- Beer AJ, Wieder HA, Lordick F, et al. Adenocarcinomas of esophagogastric junction: multi–detector row CT to evaluate early response to neoadjuvant chemotherapy. Radiology 2006;239(2):472–480.[Abstract/Free Full Text]
- Bradley JD, Ieumwananonthachai N, Purdy JA, et al. Gross tumor volume, critical prognostic factor in patients treated with three-dimensional conformal radiation therapy for non-small-cell lung carcinoma. Int J Radiat Oncol Biol Phys 2002;52(1): 49–57.[CrossRef][Medline]
- Dachman AH, MacEneaney PM, Adedipe A, Carlin M, Schumm LP. Tumor size on computed tomography scans: is one measurement enough? Cancer 2001;91(3):555–560.[CrossRef][Medline]
- Sohaib SA, Turner B, Hanson JA, Farquharson M, Oliver RT, Reznek RH. CT assessment of tumour response to treatment: comparison of linear, cross-sectional and volumetric measures of tumour size. Br J Radiol 2000;73(875):1178–1184.[Abstract]
- Yankelevitz DF, Reeves AP, Kostis WJ, Zhao B, Henschke CI. Small pulmonary nodules: volumetrically determined growth rates based on CT evaluation. Radiology 2000;217(1):251–256.[Abstract/Free Full Text]
- Choi H. Critical issues in response evaluation on computed tomography: lessons from the gastrointestinal stromal tumor model. Curr Oncol Rep 2005;7(4):307–311.[Medline]
- Belton AL, Saini S, Liebermann K, Boland GW, Halpern EF. Tumour size measurement in an oncology clinical trial: comparison between off-site and on-site measurements. Clin Radiol 2003; 58(4):311–314.[CrossRef][Medline]
- Schwartz LH, Ginsberg MS, DeCorato D, et al. Evaluation of tumor measurements in oncology: use of film-based and electronic techniques. J Clin Oncol 2000;18(10):2179–2184.[Abstract/Free Full Text]
- Van Hoe L, Van Cutsem E, Vergote I, et al. Size quantification of liver metastases in patients undergoing cancer treatment: reproducibility of one-, two-, and three-dimensional measurements determined with spiral CT. Radiology 1997;202(3): 671–675.[Abstract/Free Full Text]
- Petrou M, Quint LE, Nan B, Baker LH. Pulmonary nodule volumetric measurement variability as a function of CT slice thickness and nodule morphology. AJR Am J Roentgenol 2007;188(2):306–312.[Abstract/Free Full Text]
- Revel MP, Bissery A, Bienvenu M, Aycard L, Le-fort C, Frija G. Are two-dimensional CT measurements of small noncalcified pulmonary nodules reliable? Radiology 2004;231(2):453–458.[Abstract/Free Full Text]
- Staron RB, Ford E. Computed tomographic volumetric calculation reproducibility. Invest Radiol 1986;21(3):272–274.[CrossRef][Medline]
- Winer-Muram HT, Jennings SG, Meyer CA, et al. Effect of varying CT section width on volumetric measurement of lung tumors and application of compensatory equations. Radiology 2003;229(1): 184–194.[Abstract/Free Full Text]
- Slovis TL. The ALARA concept in pediatric CT: myth or reality? Radiology 2002;223(1):5–6.[Free Full Text]
- Therasse P, Le Cesne A, Van Glabbeke M, Verweij J, Judson I. RECIST vs. WHO: prospective comparison of response criteria in an EORTC phase II clinical trial investigating ET-743 in advanced soft tissue sarcoma. Eur J Cancer 2005; 41(10):1426–1430.[CrossRef][Medline]
- Tuma RS. Sometimes size doesn’t matter: reevaluating RECIST and tumor response rate endpoints. J Natl Cancer Inst 2006;98(18):1272–1274.[Free Full Text]
- Husband JE, Hawkes DJ, Peckham MJ. CT estimations of mean attenuation values and volume in testicular tumors: a comparison with surgical and histologic findings. Radiology 1982;144(3):553–558.[Abstract/Free Full Text]
- Shankar S, vanSonnenberg E, Desai J, Dipiro PJ, Van Den Abbeele A, Demetri GD. Gastrointestinal stromal tumor: new nodule-within-a-mass pattern of recurrence after partial response to imatinib mesylate. Radiology 2005;235(3):892–898.[Abstract/Free Full Text]
- Bruzzi P, Del Mastro L, Sormani MP, et al. Objective response to chemotherapy as a potential surrogate end point of survival in metastatic breast cancer patients. J Clin Oncol 2005;23(22):5117–5125.[Abstract/Free Full Text]
- Buyse M, Thirion P, Carlson RW, Burzykowski T, Molenberghs G, Piedbois P. Relation between tumour response to first-line chemotherapy and survival in advanced colorectal cancer: a meta-analysis. Meta-Analysis Group in Cancer. Lancet 2000;356(9227):373–378.[CrossRef][Medline]
- Goffin J, Baral S, Tu D, Nomikos D, Seymour L. Objective responses in patients with malignant melanoma or renal cell cancer in early clinical studies do not predict regulatory approval. Clin Cancer Res 2005;11(16):5928–5934.[Abstract/Free Full Text]
- Ratain MJ. Phase II oncology trials: let’s be positive. Clin Cancer Res 2005;11(16):5661–5662.[Free Full Text]
- Shepherd FA, Rodrigues Pereira J, Ciuleanu T, et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 2005;353(2):123–132.[Abstract/Free Full Text]
- Thatcher N, Chang A, Parikh P, et al. Gefitinib plus best supportive care in previously treated patients with refractory advanced non-small-cell lung cancer: results from a randomised, placebo-controlled, multicentre study (Iressa Survival Evaluation in Lung Cancer). Lancet 2005;366(9496): 1527–1537.[CrossRef][Medline]
- Coleman RE, Laymon CM, Turkington TG. FDG imaging of lung nodules: a phantom study comparing SPECT, camera-based PET, and dedicated PET. Radiology 1999;210(3):823–828.[Abstract/Free Full Text]
- Dorow DS, Cullinane C, Conus N, et al. Multi-tracer small animal PET imaging of the tumour response to the novel pan-Erb-B inhibitor CI-1033. Eur J Nucl Med Mol Imaging 2006;33(4): 441–452.[CrossRef][Medline]
- Su H, Bodenstein C, Dumont RA, et al. Monitoring tumor glucose utilization by positron emission tomography for the prediction of treatment response to epidermal growth factor receptor kinase inhibitors. Clin Cancer Res 2006;12(19):5659–5667.[Abstract/Free Full Text]
- Delbeke D. Oncological applications of FDG PET imaging. J Nucl Med 1999;40(10):1706–1715.[Free Full Text]
- Weber WA, Figlin R. Monitoring cancer treatment with PET/CT: does it make a difference? J Nucl Med 2007;48(suppl 1):36S–44S.[Abstract/Free Full Text]
- Shankar LK, Hoffman JM, Bacharach S, et al. Consensus recommendations for the use of 18F-FDG PET as an indicator of therapeutic response in patients in National Cancer Institute Trials. J Nucl Med 2006;47(6):1059–1066.[Free Full Text]
- Young H, Baum R, Cremerius U, et al. Measurement of clinical and subclinical tumour response using [18F]-fluorodeoxyglucose and positron emission tomography: review and 1999 EORTC recommendations. European Organization for Research and Treatment of Cancer (EORTC) PET Study Group. Eur J Cancer 1999;35(13):1773–1782.[CrossRef][Medline]
- Gwyther SJ. Current standards for response evaluation by imaging techniques. Eur J Nucl Med Mol Imaging 2006;33(suppl 13):11–15.[CrossRef][Medline]
- Padhani AR, Husband JE. Are current tumour response criteria relevant for the 21st century? Br J Radiol 2000;73(874):1031–1033.[Medline]
Top
Abstract
Introduction
