The role of CT in defining criteria for tumour response

Angela M Riddell,
Dow-Mu Koh, Mike King
Department of Radiology, Royal Marsden Hospital NHS
Foundation Trust, London, UK

Address for correspondence:
Dr Angela M Riddell
Department of Radiology, Royal Marsden Hospital NHS,
Foundation Trust, Fulham Road, London, SW3 6JJ, UK.
Tel: +44-(0)208-661-3964 Fax: +44-(0)208-661-3506
Email: Angela.Riddell@rmh.nhs.uk


Abstract
This article reviews the methods of assessing tumour response using CT, describing the more traditional methods (based on an alteration in tumour size) such as Response Evaluation Criteria in Solid Tumours (RECIST), and also the emerging techniques using functional imaging. Using the newer techniques of Dynamic Perfusion CT imaging and Positron Emission Tomography (PET)-CT, more detailed analysis of tumour biology can be achieved, refining our ability to assess tumour response to treatment.

Introduction
Computed tomography (CT) is currently the most widely used imaging technique for monitoring response to anti-cancer herapies. This is usually used in conjunction with tumour size measurement to determine treatment response. Until the 1980s, there was variability in the criteria applied to assess response to anti-cancer therapies, making it impossible to directly compare the response rates between different research groups developing new drugs.

The World Health Organisation (WHO) recognised the need for uniformity and devised standardised criteria for response assessment, primarily using CT as the imaging modality. This system used bi-dimensional measurements of lesions to calculate the ‘product’ of the longest diameter multiplied by its perpendicular.1 Four categories of response were described: complete response (CR), partial response (PR), stable disease and disease progression (PD). A partial response was defined as a decrease of 50% or more of the total tumour burden while disease progression was defined as an increase of 25% or more of the total tumour burden. A summary of the classification system is shown in Table 1.

RECIST (Response Evaluation Criteria in Solid Tumours)

Table 1. A comparison of the criteria for response assessment as
defined by the WHO and the RECIST study team.
Best response
 WHO: Change
in sum of
products of
bi-dimensional
measurements		 RECIST: Change
in sum of
diameter
measurements
Complete response No visible tumour No visible tumour
Partial response 50% decrease

30% decrease

Stable disease Neither PR or PD
criteria met		 Neither PR or PD
criteria met
Disease
progression
 25% increase 20% increase or
the appearance of
a new lesion
PR=partial response; PD=disease progression; WHO=World Health
Organisation; RECIST=Response Evaluation Criteria in Solid Tumours
Modifications to the WHO criteria were made in 2000, with the publication of RECIST as a result of collaboration between the European Organisation for Research and Treatment of Cancer (EORTC), the US National Cancer Institute and the National Cancer Institute of Canada Clinical Trials Group [2]. Once again, the marker for response assessment was an alteration in tumour size measured using CT, although it was acknowledged that MRI could also be used for response evaluation. The major alteration was the use of uni-dimensional measurements (longest diameter) instead of bi-dimensional measurements to quantify change in tumour size (Figure 1). A large retrospective review had shown very good concordance for overall response assessment (kappa statistic, 0.95) between uni-dimensional and bi-dimensional criteria [3]. The RECIST criteria indicated that tumour load should be represented by pre-selected measurable lesions on the baseline scan, which are set as ‘target lesions’. Sites of disease not able to be measured (e.g. ascitic fluid) should be documented and monitored as ‘non-target lesions’.

Imaging methods using RECIST
The widespread availability of CT makes it the imaging modality of choice for assessing response to treatment for patients in clinical trials. The RECIST criteria make specific recommendations regarding the imaging protocols which should be used for CT and, in particular, stipulate that the diameter of target lesions should be no less than double the CT slice thickness to reduce error related to the ‘partial volume’ effect. The majority of oncology imaging centres are now equipped with spiral CT scanning machines and reconstruct images contiguously at 5 mm intervals, allowing for the target lesions to be a minimum of 10 mm in size. MRI can be used to monitor response to treatment but the variability in equipment and protocols results in difficulties when comparing response evaluation between centres. The RECIST criteria stipulate that ultrasound should not be used in clinical trials unless the lesion is superficial. In all cases, the same imaging modality should be used throughout the study and, for CT, the target lesions should be measured on the same window settings to maintain uniformity.

The left inguinal lymph node demonstrates disease progression both by the bi-dimensional WHO criteria: >25% increase in the sum of the diameters (a, b) and the RECIST uni-dimensional criteria: >20% increase in the longest diameter (c,d)
Figure 1. The left inguinal lymph node demonstrates disease progression both by the bi-dimensional WHO criteria: >25% increase in the sum of the diameters (a, b) and the RECIST uni-dimensional criteria: >20% increase in the longest diameter (c,d).


Assessment of response using RECIST
  Diameter,
2r		 Product,
(2r) [2]		 Volume,
4/3pr [3]
Response Decrease Decrease Decrease
30% 50% 65%
50% 75% 87%
Disease
Progression		 Increase Increase Increase
12% 25% 40%
20% 44% 73%
25% 56% 95%
30% 69% 120%
Table 2. Evaluation of response and disease progression using RECIST recommendations.
Using RECIST, the main four response categories remain as described for the WHO criteria, although with modifications to the relationship between a change in the diameter (2r), product (2r) [2] and volume (4/3p r [3]) of a sphere (Table 2). A partial response is defined as a decrease in the sum of the longest diameters of all target lesions by >=30%. Disease progression is defined by an increase of >=20% of the sum of the longest diameters of all target lesions, or by the appearance of a new lesion (Table 1).

RECIST recommends that up to a maximum of 5 target lesions per organ and 10 lesions in total to be measured and recorded at baseline. The presence of non-target lesions should be documented at baseline and their presence or absence recorded throughout follow-up Evaluation of response within RECIST is based on the best overall response from baseline and the status of PR or CR requires confirmation on subsequent follow-up at least 4 weeks after the criteria for response are met.

A recent study assessing the response of colorectal liver metastases to chemotherapy has challenged the requirement to measure five target lesions, as defined by the RECIST criteria [4].For a study group of 30 patients, the results showed that measurement of the maximal diameter of the single largest liver lesion yielded the same treatment-response classification as measuring up to five lesions. This indicates that it may be possible to reduce the number of lesions measured in the liver for future clinical trials. However, multiple measurements are still recommended for assessing some diseases such as malignant mesothelioma. A study demonstrated that the sum of six measurements - two taken from different positions at three different levels on thoracic CT imaging - could be taken as a uni-dimensional measurement when applying RECIST criteria [5]. This modified approach has been shown to correlate with patient outcome and pulmonary function test.

Limitations to RECIST
Although the RECIST criteria remain a robust technique for monitoring response and continue to be used as primary or secondary end-points for clinical trials, there are recognised limitations to the effectiveness of using these criteria. The assessment of response is based purely on alteration in tumour size and does not assess the biological function of the tumour. Errors in measurement can occur due to difficulty in differentiation between viable tumour and fibrosis and in delineating poorly marginated or infiltrative tumours. In addition, a change in the CT tumour characteristics is not recognised by RECIST criteria. For example, while a tumour may not change significantly in size, it may become low density or show reduced contrast enhancement. It should be noted that necrotic lesions are considered to be ‘nontarget’ lesions. In bladder cancers, calcifications within tumours may also occur as a response to chemotherapy [6]. However, these observations are not accepted as objective measures of disease responses.

Functional imaging
Functional imaging may be performed using CT, magnetic resonance imaging (MRI), positron emission tomography (PET), and more recently PET-CT. By employing special scanning techniques or tracer substances, changes observed at imaging are linked to changes in tumour pathophysiology. Functional imaging techniques provide information that reflects the tumour micro-environment, such as tumour vascularity, cellularity, metabolism, hypoxia and acidosis.

The development of functional imaging in the past decade has been driven in part by the limitations of current measurement criteria, but more importantly, by the introduction of molecular targeted therapies. Novel cancer therapeutics are frequently targeted against specific biological targets, such as cellular receptors and gene expression. These types of treatments may not result in a significant reduction in tumour size. An example of this would be the use of an anti-vascular drug that disrupts and destroys tumour blood vessels, without which a tumour cannot grow. In some instances, there may even be a paradoxical increase in tumour size despite successful treatment [7].

New functional imaging markers derived using CT, MRI and PET are being widely evaluated in vivo and in vitro, especially in the field of drug development. Functional imaging is being applied to determine the biological active dose of novel therapeutics and also for the monitoring of treatment effects. There is a growing body of evidence to suggest that some of these techniques can detect changes related to treatment as early as 24 hours following anti-cancer treatment. The degree of early change in functional imaging measurements has been shown to be predictive of response by RECIST criteria at the end of treatment [8-13]. However, many of these techniques have not yet translated into routine clinical use because they require considerable physics expertise for image acquisition and processing, which is not widely available.

Functional CT Imaging
Figure 2
Figure 2. (a) Conventional CT image and maps of (b) blood volume, (c) mean transit time and (d) permeability-surface area product. There is a heterogeneously enhancing lymph node (circled) seen on the morphological image measuring 2 cm in size. Areas with relatively less enhancement within the node correspond to regions of low blood flow and lower permeability on the functional parametric maps. These parametric maps provide complementary information related to tumour vascularity. (Courtesy of Dr. K Newbold/A Sohaib)

Dynamic perfusion CT imaging is a technique used to assess and measure tumour vascularity. Following a rapid intravenous injection of contrast medium, repeated scanning of a specific tumour region is performed, typically one image every few seconds, to track the passage of contrast through the tumour tissue. By drawing a region of interest over the tumour on the CT image, the change in the CT attenuation value of the tumour with time, due to contrast passage, can be charted. This CT attenuation-time curve can be visually assessed for characteristics of tumour circulation. Tumour tissues typically show increased vascularity with rapid onset and early enhancement peak compared with normal tissue. This is followed by early contrast washout from the tumour.

A mathematical model can be fitted to the CT attenuation-time curve to derive quantitative parameters which describe the tumour microcirculation. For example, the permeability-surface area product describes the rate of contrast leakage into the extracellular space, which is increased in tumour tissue. This results from hyperpermeable tumour vessels arising from angiogenesis. The permeability-surface area product has been shown to decrease following initiation of anti-vascular treatment (Figure 2). Other quantitative parameters that are derived by such mathematical modelling include the tumour blood volume, tumour blood flow and the mean transit time. Dedicated software to perform these analyses are now becoming more widely available on standard CT workstations.

Experience is growing in the application of perfusion CT for tumour assessment. One recent study of perfusion CT in patients with rectal cancer [13] performed before and after chemoradiation treatment, demonstrated significant reduction in the tumour blood volume and an increase in the mean transit time in patients who responded to treatment. In the same study, it was found that patients with initial high tumour blood volume and short mean transit time were associated with poor response.

The main disadvantage of perfusion CT imaging is the potential radiation burden as a result of these studies. This imposes a limit on the volume of the body that can be evaluated, the temporal resolution of imaging, and the frequency with which the measurements can be repeated. Not surprisingly, many functional CT studies have employed only single or a few imaging sections through an area of interest to minimise radiation exposure.

PET-CT imaging

Figure 3. Staging 18F-FDG PET/CT (right) in a patient with non- Hodgkin’s lymphoma showing bulky nodal disease encasing the left shoulder and proximal humerus, with left supra-, infra-clavicular fossa and axillary nodes (inverse grey scale images top row, CT image middle row and fused colour scale PET/CT images bottom row). Serial follow-up PET/CT (right to left) shows an excellent metabolic response to treatment. However, significant residual soft tissue remained on CT imaging (partial response) but there was significantly less metabolic activity observed at PET imaging. (Courtesy of Dr. B Sharma)

In recent years, PET imaging has changed the landscape of oncological imaging. Imaging is achieved by labelling molecules with a positron-containing moiety. When these compounds are administered, the positrons are annihilated by collision with electrons in the body, resulting in energy release that is captured by the scan detectors. The strength of this imaging technique is that, depending on the choice of the radio-labelled compound, a specific metabolic pathway, protein, receptor or gene expression may be interrogated. For example, the most common radio-labelled tracer used for imaging is 18F-fluoro- deoxyglucose (18F-FDG). This is a glucose analogue that is transported into cells but becomes trapped by phosphorylation. Increased tracer activity is usually observed in tumour tissue, which has higher glucose metabolism when compared with normal tissue.

The spatial resolution of PET imaging alone is relatively poor by comparison with CT or MR imaging. Furthermore, the specificity of the tracer means that there may be little activity in normal tissue, leading to poor visualisation of normal anatomical structures. The advent of PET-CT imaging by combining functional information obtained by PET with high spatial resolution anatomical information derived from concurrent CT study, has overcome the main shortcomings of PET imaging alone.

PET-CT is now widely used in oncology for cancer staging as well as for the assessment of treatment response. In patients with lymphoma, conventional CT and MR imaging cannot discriminate between active disease and fibrosis/necrosis within residual nodal masses, which frequently persist despite successful treatment. PET imaging is now recognised as the most accurate method in distinguishing between residual disease and post-treatment change in patients with non-Hodgkin’s lymphoma [14,15]. Using 18F-FDG-PET, successful treatment results in reduction or disappearance of metabolic tracer activity in nodal tissue [16,17].

Conclusions
CT imaging with the application of size criteria is still the most widely used method for assessing tumour response. However, functional imaging techniques using CT are becoming increasing important for providing unique information, which inform therapeutic decisions and management strategies.

Key Learning
  • The RECIST criteria are currently the most widely used method for assessing treatment response. The marker for response is an alteration in the longest diameter of the tumour. CT is the imaging modality most frequently used for response evaluation
  • The RECIST criteria limit response assessment to an alteration in tumour size, with no facility to measure an alteration in biological function. Post-treatment CT measurements can be inaccurate due to the inability to differentiate fibrosis from residual viable tumour
  • Functional imaging techniques using either PET-CT or CT perfusion analysis enable quantification of tumour metabolic activity and vascularity. These parameters allow for alterations to the biological function of the tumour to be calculated as part of the evaluation of treatment respons
References
  1. Miller AB, Hoogstraten B, Staquet M, et al . Reporting results of cancer treatment. Cancer 1981;47(1):207-14.
  2. 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):205-16.
  3. James K, Eisenhauer EA, Christian M, et al . Measuring response in solid tumors: unidimensional versus bidimensional measurement. J Natl Cancer Inst 1999;91(6):523-8.
  4. Zacharia TT, Saini S, Halpern EF, et al . 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):1067-70.
  5. Byrne MJ, Nowak AK. Modified RECIST criteria for assessment of response in malignant pleural mesothelioma. Ann Oncol 2004;15(2):257-60.
  6. Dore B, Aubert J. Re: Bladder wall calcification after intravesical mitomycin C treatment of superficial bladder cancer. J Urol 1991;145(6):1277.
  7. Choi H. Critical issues in response evaluation on computed tomography: lessons from the gastrointestinal stromal tumor model. Curr Oncol Rep 2005;7(4):307-11.
  8. Moffat BA, Chenevert TL, Lawrence TS, et al . Functional diffusion map: a noninvasive MRI biomarker for early stratification of clinical brain tumor response. Proc Natl Acad Sci U S A 2005;102(15):5524-9.
  9. Mardor Y, Roth Y, Ochershvilli A, et al . Pretreatment prediction of brain tumors' response to radiation therapy using high b-value diffusionweighted MRI. Neoplasia 2004;6(2):136-42.
  10. Dugdale PE, Miles KA, Bunce I, et al . CT measurement of perfusion and permeability within lymphoma masses and its ability to assess grade, activity, and chemotherapeutic response. J Comput Assist Tomogr 1999;23(4):540-7.
  11. Gandhi D, Chepeha DB, Miller T, et al . Correlation between initial and early follow-up CT perfusion parameters with endoscopic tumor response in patients with advanced squamous cell carcinomas of the oropharynx treated with organ-preservation therapy. AJNR Am J Neuroradiol 2006;27(1):101-6.
  12. Prokop, M. CT and MR to assess the response of liver tumors to hepatic perfusion. Recent Results Cancer Res 1998;147:136-54. 13. Sahani DV, Kalva SP, Hamberg LM et al . Assessing tumor perfusion and treatment response in rectal cancer with multisection CT: initial observations. Radiology 2005;234(3):785-92.
  13. Guermazi A, Juweid ME. Commentary: PET poised to alter the current paradigm for response assessment of non-Hodgkin's lymphoma. Br J Radiol 2006;79(941):365-7.
  14. Kostakoglu L, Goldsmith SJ. 18F-FDG PET evaluation of the response to therapy for lymphoma and for breast, lung, and colorectal carcinoma. J Nucl Med 2003;44(2):224-39.
  15. Juweid ME, Cheson BD. Role of positron emission tomography in lymphoma. J Clin Oncol 2005;23(21):4577-80.
  16. Juweid ME, Wiseman GA, Vose JM, et al . Response assessment of aggressive non-Hodgkin's lymphoma by integrated International Workshop Criteria and fluorine-18-fluorodeoxyglucose positron emission tomography. J Clin Oncol 2005;23(21):4652-61.


10-2006 BUY1145315/JB2356/MB002264/CMC 12th edition