Tumor Response in Patients With Advanced Non Small Cell Lung Cancer: Perfusion CT Evaluation of Chemotherapy and Radiation Therapy

Transcription

1 Cardiopulmonary Imaging Original Research Wang et al. CT of Non Small Cell Lung Cancer Cardiopulmonary Imaging Original Research Jianwei Wang 1 Ning Wu 1 Matthew D. Cham 2 Ying Song 1 Wang J, Wu N, Cham MD, Song Y Keywords: chemotherapy, lung neoplasms, perfusion CT, prognosis, radiation therapy DOI: /AJR Received June 9, 2008; accepted after revision March 4, Department of Diagnostic Radiology, Cancer Hospital and Institute, Chinese Academy of Medical Sciences, Peking Union Medical College, PO Box 2258, 17 Panjiayuan Nanli, Beijing , China. Address correspondence to N. Wu. 2 Department of Radiology, New York Presbyterian Hospital, Weill Medical College of Cornell University, New York, NY. AJR 2009; 193: X/09/ American Roentgen Ray Society Tumor Response in Patients With Advanced Non Small Cell Lung Cancer: Perfusion CT Evaluation of Chemotherapy and Radiation Therapy OBJECTIVE. The objectives of this study were to prospectively evaluate changes in tumor perfusion after chemoradiation therapy and to investigate the feasibility of perfusion CT for prediction of early tumor response and prognosis of non small cell lung cancer. SUBJECTS AND METHODS. Perfusion CT was performed on an MDCT scanner with 50 ml of iodinated contrast material injected at 4 ml/s. The quality of each functional map was rated from 0 to 3 for 123 patients with confirmed lung cancer. A subset of images was independently reviewed by two radiologists to measure interobserver and intraobserver variability. Perfusion parameters and tumor response were assessed for 35 patients with non small cell lung cancer who underwent chemoradiation therapy. Progression-free survival and overall survival were analyzed for 22 patients who underwent repeated perfusion CT after therapy. RESULTS. Image quality was graded 2 (moderate) or 3 (good) in 68.2% of cases. High interobserver and intraobserver correlations of perfusion parameters were found on qualified images. The patients who responded to chemoradiation therapy had significantly greater blood flow (p = 0.023) than patients who did not respond. The median progression-free survival period of the patients with an increased permeability surface area product was 4.7 months, significantly lower than the median progression-free survival period of 19.0 months among patients with a decreased permeability surface area product (p < 0.001). The median overall survival period was 10.6 months for the group with an increased permeability surface area product, significantly lower than the 19.3 months for the group with a decreased permeability surface area product (p = 0.004). CONCLUSION. Non small cell lung cancer with higher perfusion is more sensitive to chemoradiation therapy than that with lower perfusion. After chemoradiation therapy, findings at perfusion CT are a significant predictor of early tumor response and overall survival among patients with non small cell lung cancer. T he Response Evaluation Criteria in Solid Tumors (RECIST) were published in February 2000 and became widely applicable in clinical trials [1]. The RECIST were evaluated in several articles and recommended for tumor evaluation in future trials of therapy for non small cell lung cancer (NSCLC) [2 6]. Although they are useful for estimating tumor response to therapies based on lesion size, the RECIST have disadvantages. First, response assessment with morphologic imaging techniques such as CT has limitations in reliable differentiation of necrotic tumor or fibrotic scar from residual tumor tissue. Second, the objective response rates for various platinum-based regimens are in the range of 20 40% for advanced NSCLC [7, 8]. Because of the relatively slow tumor shrinkage, as measured on CT images, a substantial percentage of patients may undergo several weeks of toxic therapy without clinical benefit. Thus early prediction of tumor response is of particular importance to patients with advanced NSCLC. Furthermore, some of the new targeted therapies are cytostatic rather than cytoreductive, in which case successful treatment does not actually reduce tumor size, posing new demands on imaging techniques [9]. FDG PET or PET/CT after initiation of chemotherapy, radiation therapy, or both for NSCLC can be used to assess tumor response to therapy by depicting a reduction in the metabolic activity of the tumor, which is a favorable prognostic indicator of survival [9, 10]. Several studies [11 13] have shown that perfusion CT has potential for monitoring the 1090 AJR:193, October 2009

2 CT of Non Small Cell Lung Cancer effects of chemotherapy with or without radiation therapy and for predicting the response of rectal and oropharyngeal cancer. To the best of our knowledge, there are no data on the use of perfusion CT for monitoring the response of NSCLC to chemotherapy. The cost-to-benefit ratio of perfusion CT relative to PET and PET/CT for advanced NSCLC is unknown. The purpose of our study was to prospectively monitor changes in tumor perfusion after chemotherapy, radiation therapy, or both and to determine whether perfusion parameters can be used for prognosis and prediction of early tumor response to therapy for NSCLC. Subjects and Methods Patient Inclusion Criteria The study was approved by our institutional review board, and written informed consent was obtained from all participants. From April 1, 2005, to July 31, 2006, all patients with suspected advanced lung cancer at our hospital were offered the opportunity to undergo perfusion CT if they met the following inclusion criteria: longest diameter of the tumor of 3.0 cm or greater and tumor believed to be a T3 or T4 lesion with or without nodal involvement or metastatic disease, no previous chemotherapy or radiation therapy, normal pulmonary function without asthma or dyspnea, body weight of kg, and no contraindication to iodinated contrast medium. Among 152 patients undergoing perfusion CT during this period, 123 patients had lung cancer pathologically proven with surgery, bronchoscopy, or biopsy. Twentyfive patients did not have histologic confirmation, and four patients had tuberculosis (n = 3) or sarcoidosis (n = 1). Perfusion CT Techniques All patients were trained to hold their breath at end-inspiration for 50 seconds during perfusion CT. Patients unable to keep holding their breath toward the end of the 50-second acquisition were asked to take very small breaths. Perfusion CT was performed with a 16-MDCT (LightSpeed Pro 16, GE Healthcare) or an 8-MDCT (LightSpeed Ultra, GE Healthcare) scanner. A preliminary unenhanced CT scan was obtained to localize the tumor at a slice thickness of 5 mm. A 20-mm scanning range for perfusion CT covered the maximum tumor area. Dynamic imaging of this volume was performed with the following parameters: four contiguous 5-mm sections at the same table position, 1-second gantry rotation time, 120 kvp, 50 ma, and 50-second acquisition time. Fifty milliliters of nonionic iodinated contrast medium (iohexol 300 mg I/mL, Omnipaque, GE Healthcare) was injected at a rate of 4 ml/s through the antecubital vein. The scanning delay was 10 seconds. After the dynamic study another 50 ml of contrast medium was injected at a rate of 3 ml/s for contrast-enhanced chest CT from the thoracic inlet through the adrenal glands at mm slice thickness, 13.5 mm/s table speed, 120 kvp, 200 ma, and no scan delay. Image and Data Analysis All images were reviewed on a workstation (Advantage Windows 4.2, GE Healthcare). The image quality of each functional map was rated on a 4-point scale by a radiologist with 13 years of experience in chest CT and 1 year of experience in perfusion CT. The image quality was classified as grade 0, failed, if functional maps could not be obtained; grade 1, poor, if there was severe respiratory misregistration or other artifact; grade 2, moderate, for minimal respiratory misregistration or other artifacts; and grade 3, good, for no respiratory misregistration or other artifacts. Grade 2 and 3 images were classified as qualified images and analyzed by a second radiologist, who had 1 year of experience in chest CT and 2 months of experience in perfusion CT. The two readers were blinded to each other s results to allow measurement of interobserver variability. For measurement of intraobserver variability, the more experienced reader repeated the perfusion CT analysis on the same data set 7 12 months after the first reading. Perfusion parameters were calculated with the manufacturer s software (CT Perfusion 3, GE Healthcare). The attenuation threshold was defined between 20 and 300 HU to increase the speed of calculation. The arterial input was obtained from a 2- to 6-pixel region of interest placed in the descending aorta, the aortic arch, or main branches of the aorta according to the location of the tumor. A central region of tumor was chosen, and an elliptic or freehand region of interest was drawn around the tumor. The region of interest excluded necrotic areas, regions of atelectatic lung, vessels, and calcifications. Functional maps were generated with color scales based on the following perfusion parameters: blood flow per 100 grams of wet tissue per minute; blood volume per 100 grams of wet tissue; mean transit time of contrast material in the local vascular system; and permeability surface area product, which is the rate of contrast leakage into the extracellular space. All the images were saved in a PACS as the part of the patients permanent medical records. Clinical Treatment and Follow-Up Patients with inoperable NSCLC were treated with chemotherapy, radiation therapy, or concurrent chemoradiotherapy. Platinum-based first-line chemotherapeutic regimens were used. The radiation therapy plan consisted of 60 Gy in 30 fractions of 2 Gy each. The follow-up perfusion CT scan was obtained with the same protocol as the baseline perfusion CT scan after two-cycle chemotherapy or before the end of radiation therapy (when the total dose reached Gy). At the same time, tumor response was evaluated according to the RECIST criteria [1], including target lesions and nontarget lesions at contrast-enhanced CT. Tumor response after therapy was classified as complete response, the disappearance of all target lesions; partial response, at least a 30% decrease in the sum of the longest diameters of target lesions, the reference being the baseline sum of the longest diameters; stable disease, neither sufficient shrinkage to qualify for partial response nor sufficient increase to qualify for progressive disease, the reference being the smallest sum of the longest diameters since treatment started; or progressive disease, at least a 20% increase in the sum of the longest diameters of target lesions, the reference being the smallest sum of the longest diameters recorded since treatment started or the appearance of one or more new lesions. Patients with a complete response or partial response were further classified as responders; the patients with stable disease or progressive disease were further classified as nonresponders. During and after treatment, patients underwent clinical evaluations at regular intervals. The evaluations included chest CT, radiography, and other imaging studies as indicated. The follow-up closeout date was March 31, No patients were lost to follow-up. The median follow-up time was 13.7 months (range, months). The date of local or distant progression was defined as the earliest date at which disease progression was confirmed clinically, with imaging, or through biopsy. Progression-free survival was measured from the baseline perfusion CT scan to the date of progression or death of any cause. Overall survival was measured from the date of the baseline perfusion CT scan to the date of death. Patients who were alive on the closeout date had their survival data censored to that date. Statistical Analysis The Kolmogorov-Smirnov test was used to determine whether the data fit a normal distribution. Interobserver and intraobserver agreement on perfusion CT analysis was assessed with the Bland-Altman method [14]. The mean difference, SD of the differences, and 95% limits of agreement (mean difference 2SD and mean difference + 2SD) were calculated for each of the four perfusion parameters. AJR:193, October

3 Wang et al. Baseline perfusion parameters of NSCLC were compared between the responders and nonresponders by use of independent samples Student s t tests. Perfusion parameters before and after therapy were compared by use of the paired samples Student s t test. Progression-free survival and overall survival with respect to the parameter changes were calculated with Kaplan-Meier estimates. SPSS software (version 16.0, SPSS) for Microsoft Windows was used for all data analysis except the Bland- Altman test, which was performed with the Graph- Pad Prism 5 program (GraphPad Software). Statistical tests were based on a two-sided significance level set at Results Image Quality The image quality of the functional maps of 123 patients with pathologically proven lung cancer was rated on the 4-point scale as follows: grade 0, n = 20 (16.3%); grade 1, n = 19 (15.4%); grade 2, n = 42 (34.1%); grade 3, n = 42 (34.1%). The most common cause of perfusion CT failure (grade 0) was input artery enhancement before the start of the CT acquisition (n = 11). A second cause of failure was inadequate perfusion CT coverage of the maximum tumor area owing to respiratory movement (n = 5). Failure also occurred when perfusion CT covered only a region of lymphadenopathy (n = 4). Interobserver and Intraobserver Agreement The grade 0 and grade 1 cases were excluded from analysis. The grade 2 (n = 42) and grade 3 (n = 42) cases were used to assess interobserver and intraobserver agreement. Tables 1 and 2 summarize the mean value, SD, mean difference, and 95% limits of agreement for the two paired sets of tumor measurements. Patient Characteristics Among the 84 patients who had qualified images (grades 2 and 3), 57 patients underwent chemotherapy, radiation therapy, or concurrent chemoradiotherapy, and the other 27 underwent lobectomy or pneumectomy. Response assessment after two cycles of chemotherapy or before the end of radiation therapy was successfully performed on 44 patients, including 35 patients with NSCLC and nine patients with small cell lung cancer (SCLC). The nine patients with SCLC were excluded from this study. The characteristics of the 35 patients with NSCLC are summarized in Table 3. After follow-up perfusion CT, 13 patients were excluded for the following reasons: TABLE 1: Interobserver Agreement for Tumor Measurements With Acceptable Image Quality (n = 84) Perfusion Measurement Blood flow (ml/min/100 g) grade 0 and 1 image quality (n = 8), declining to undergo follow-up perfusion CT (n = 3), and lack of adherence to the response assessment plan (n = 2). Baseline Perfusion Parameters in Responders and Nonresponders Among the 35 patients with NSCLC who underwent response assessment, 21 patients were classified as responders, including nine patients treated with chemotherapy, six patients treated with radiation therapy, and six patients treated with concurrent chemoradiotherapy. Fourteen patients were classified as nonresponders, including 10 patients Difference 95% Limits of Agreement Observer (31.157) , Observer (29.727) Blood volume (ml/100 g) Observer (1.658) , Observer (1.682) transit time (s) Observer (3.909) , Observer (3.809) Permeability surface area product (ml/min/100 g) Observer (6.492) , Observer (5.768) Note Values in parentheses are SD. TABLE 2: Intraobserver Agreement for Tumor Measurements With Acceptable Image Quality (n = 84) Blood flow (ml/min/100 g) Perfusion Measurement Difference 95% Limits of Agreement First (32.514) , Second (29.727) Blood volume (ml/100 g) First (1.729) , Second (1.682) transit time (s) First (3.934) , Second (3.809) Permeability surface area product (ml/min/100 g) First (6.013) , Second (5.768) Note Values in parentheses are SD. treated with chemotherapy, one patient treated with radiation therapy, and three patients treated with concurrent chemoradiotherapy. The baseline blood flow and blood volume of responders were significantly higher than those of nonresponders (p = 0.023) (Figs. 1 and 2). No significant difference was found for blood volume (p = 0.097), mean transit time (p = 0.159), or permeability surface area product (p = 0.961) (Table 4). Perfusion Parameters Before and After Therapy in Responders and Nonresponders Among the 22 patients undergoing followup perfusion CT after therapy, no significant 1092 AJR:193, October 2009

4 CT of Non Small Cell Lung Cancer TABLE 3: Patient Characteristics and Clinical Tumor Response Assessment (n = 35) Age (y) Characteristic Value 58.5 Range Sex Men 29 Women 6 Histologic finding Squamous cell carcinoma 17 Adenocarcinoma 18 Tumor stage IIB 1 IIIA 10 IIIB 10 IV 14 Therapy Chemotherapy 19 Radiation therapy 7 Concurrent radiochemotherapy 9 Responders Complete response 0 Partial response 21 Nonresponders Stable disease 10 Progressive disease 4 Note Except for age, values are numbers of patients. TABLE 4: Correlation Between Baseline Perfusion Parameters and Therapeutic Response Parameter differences were found between baseline and posttherapy measurements in mean blood flow, blood volume, mean transit time, or permeability surface area product. Among 10 patients treated with radiation therapy and con current chemoradiotherapy, followup findings showed seven responders had significant decreases in blood flow (70.9 ± 20.7 ml/min/100 g to 33.6 ± 14.3 ml/min/100 g, t = 3.781, p = 0.009) and blood volume (5.0 ± 1.2 ml/100 g to 2.4 ± 0.8 ml/100 g, t = 2.483, p = 0.048), a significant increase in mean transit time (9.0 ± 2.4 seconds to 12.4 ± 5.0 seconds, t = 4.420, p = 0.004), and a decrease in permeability surface area product that was not significant (12.7 ± 3.9 ml/min/100 g to 6.8 ± 3.7 ml/min/100 g, t = 1.600, p = 0.161). Among the responders, the only patient who had an increase in blood flow, blood volume, and permeability surface area product at follow-up also was found to have an early relapse after 5.9 months. Two of three nonresponders had increased blood flow, blood volume, and permeability surface area product and decreased mean transit time at follow-up. Among 12 patients treated with chemotherapy, increased or decreased perfusion parameters after therapy were found in both responders and nonresponders. Prediction of Survival With Perfusion CT Findings The 22 patients undergoing follow-up perfusion CT were divided into two groups according A Fig year-old man with highly perfused squamous cell carcinoma (moderately differentiated) in right middle lobe who had good response to chemotherapy. A, Functional map of blood flow before chemotherapy shows mean tumor blood flow of ml/min/100 g. B, Contrast-enhanced chest CT scan after two cycles of platinum-based chemotherapy shows marked tumor reduction (partial response). A Fig year-old man with poorly perfused adenocarcinoma (moderately differentiated) in left upper lobe who had no response to chemotherapy. A, Functional map of blood flow before chemotherapy shows mean tumor blood flow of 20.3 ml/min/100 g. B, Contrast-enhanced chest CT after two cycles of platinum-based chemotherapy shows interval enlargement of tumor (progressive disease). Responders (n = 21) Nonresponders (n = 14) t p Blood flow (ml/min/100 g) 81.0 (33.6) 56.3 (23.1) Blood volume (ml/100 g) 5.6 (1.9) 4.6 (1.2) transit time (s) 8.1 (3.9) 10.4 (5.7) Permeability surface area product (ml/min/100 g) 15.2 (5.6) 15.1 (5.4) Note Values other than t and p are mean with SD in parentheses. Data fit a normal distribution calculated with Kolmogorov-Smirnov test. Independent samples Student s t tests were used to determine t and p. to the changes in permeability surface area product after therapy. In the group with increased permeability surface area pro duct (n = 11), the mean permeability surface area product was 19.7 ± 8.1 ml/min/100 g after therapy and 13.8 ± 6.4 ml/min/100 g before therapy (Fig. 3). In the group with decreased permeability surface area product (n = 11), the mean product was 8.2 ± 5.5 ml/min/100 g after therapy and 15.3 ± 5.1 ml/min/100 g before therapy (Fig. 4). No significant differences between the groups were found in blood flow, blood volume, or mean B B AJR:193, October

5 Wang et al. transit time. The median progression-free survival period for the group with an increased permeability surface area product was 4.7 months, and that for the group with a decreased permeability surface area product was 19.0 months (log rank chi square, ; p < 0.001) (Fig. 5). The median overall survival periods were 10.6 and 19.3 months, respectively (log rank chi square, 8.154; p = 0.004) (Fig. 6). Both differences were statistically significant. Discussion Although results of a few studies on perfusion CT for lung cancer have been published, the utility of functional maps has not been well established [15, 16]. Respiratory motion has a greater effect on perfusion CT of the chest than it does on perfusion CT of the brain, neck, and pelvis, potentially reducing image quality and examination reliability. In our study the image quality of functional maps was unsatisfactory in 15.5% of patients. The major factor contributing to reduced image quality was misregistration artifact from respiratory motion. It is difficult for patients to hold their breath for 50 seconds, despite breath-hold training before perfusion CT. Use of a 64- or 320-MDCT scanner can improve misregistration through greater tumor volume coverage. Respiration-gated perfusion CT is another potential solution to misregistration artifacts. Perfusion CT failed in 16.3% of our patients. The most common cause of failure was enhancement of the input artery before the start of perfusion CT acquisition. This problem may be caused by individual differences between patients or by technologist error. We selected a reasonable scan delay of 10 seconds for our patients, although a shorter scan delay may be needed for some patients. A B C Fig year-old woman with squamous cell carcinoma (poorly differentiated) in right upper lobe, increase in permeability surface area product after radiation therapy, and early relapse of tumor. A, Functional map of permeability surface area product before radiation therapy shows mean value of 11.4 ml/min/100 g. B, Functional map of permeability surface area product after radiation therapy shows tumor shrinkage but with increased mean tumor permeability surface area product of 14.6 ml/min/100 g. C, Follow-up contrast-enhanced chest CT scan 4 months after treatment shows interval enlargement of tumor. In studies by Ng et al. [16] and Goh et al. [17 20] a 5-mL/s flow rate and 5-second scan delay were used for imaging of patients with lung cancer and colorectal cancer. We found high interobserver and intraobserver agreement in perfusion parameter measurements. The mean differences in some parameters were lower than previously reported [17, 20, 21], possibly because of the userfriendliness of our software application and the standardization of our measurement protocol, inclusion of only qualified (grade 2 and grade 3) images during analysis, and the larger sample size compared with previous studies. Our results show that baseline blood flow was significantly higher in responders than in nonresponders. This finding suggests that perfusion CT can be used to predict tumor response in patients with advanced NSCLC. Our results also show that when blood flow was greater than 100 ml/min/100 g and blood volume was greater than 6 ml/100 g, the tumor response rate was 7/7. To the best of our knowledge, there are no published data on perfusion CT for assessment of tumor response to chemotherapy for NSCLC. Bellomi et al. [11] evaluated perfusion parameters in 24 patients with rectal carcinoma. They found that the baseline blood flow and blood volume in the seven patients who did not respond to therapy were significantly lower than those of the 17 responders. Hermans et al. [22, 23] evaluated perfusion CT of 105 squamous cell carcinomas of the head and neck and found that patients with lower perfusion values had a significantly higher local failure rate. The results of these earlier perfusion CT studies on different tumors and the results of our study suggest that compared with poorly perfused tumors, highly perfused tumors may be more easily accessed with chemotherapy. Higher perfusion may allow greater oxygenation and thus potentially greater radiosensitivity. Bellomi et al. [11] observed a significant decrease in blood flow and blood volume after neoadjuvant chemoradiotherapy in the care of 19 patients with rectal carcinoma. The permeability surface area product also decreased significantly after therapy. Another study [13] on perfusion CT for monitoring neoadjuvant chemoradiotherapy in rectal cancer showed a significant decrease in blood flow and increase in mean transit time after treatment. Unlike those investigators, we found no significant differences between baseline and posttherapeutic measurements in mean blood flow, blood volume, mean transit time, or permeability surface area product in patients with NSCLC. However, in 10 patients treated with radiotherapy with or without chemotherapy, we found a correlation between tumor response and a decrease in tumor perfusion. It has been reported [24] that microvascular damage is a key mechanism in tumor response to radiation. Therefore, reduced volume of the vascular bed and leakage from neoplastic vessels due to microvascular damage after radiation therapy would be reflected by blood flow, blood volume, and permeability surface area product. Ng et al. [25] found that hypofractionated radiotherapy increased tumor vascular blood volume and permeability of NSCLC, a finding quite different from ours. The difference in findings may be due to differences in tumor vascular effects after hypofractionated and conventional radiotherapy. We also found that the only responder with an increase in blood flow, blood volume, and 1094 AJR:193, October 2009

6 CT of Non Small Cell Lung Cancer A C Fig year-old man with advanced adenocarcinoma (moderately differentiated), atelectasis in left upper lobe, decrease in permeability surface area product after radiation therapy, and stable disease for several months. A, Functional map of permeability surface area product before radiation therapy shows mean value of 13.3 ml/ min/100 g. B, Functional map of permeability surface area product after radiation therapy shows tumor shrinkage and decreased mean tumor permeability surface area product of 4.4 ml/min/100 g. C, Follow-up contrast-enhanced CT scan shows findings 1 month after therapy. D, Follow-up contrast-enhanced CT scan 3 months after treatment shows stable reduction in tumor bulk. Probability of Progression-Free Survival Progression-Free Survival (mo) Fig. 5 Graph shows results of Kaplan-Meier analysis of relation between progression-free survival period and changes in permeability surface area product before and after therapy (p < 0.001). Solid line indicates decreased permeability surface area product group (n = 11); dashed line, increased permeability surface area product group (n = 11). Probability of Overall Survival Overall Survival (mo) Fig. 6 Graph shows results of Kaplan-Meier analysis of relation between overall survival and changes in permeability surface area product before and after therapy (p = 0.004). Solid line indicates decreased permeability surface area product group (n = 11); dashed line, increased permeability surface area product group (n = 11). permeability surface area product at followup perfusion CT had an early relapse 5.9 months after treatment. This finding may suggest that such responders need additional chemotherapy or targeted therapy with increasing perfusion after radiation therapy. B D The published data on changes in perfusion parameters before and after chemotherapy are scarce [15]. Our results suggest that there is no correlation between tumor response and tumor perfusion changes in patients treated with chemotherapy, possibly because cytotoxic drugs may not affect tumor microvasculature. Additional studies are needed to explore the changes in perfusion parameters before and after targeted therapies. There is little information on the use of perfusion CT for predicting prognosis in lung cancer. We found that a decrease in permeability surface area product (mean, 7.1 ml/min/100 g) was associated with a longer median progression-free survival time (19.0 months) and a longer median overall survival time (19.3 months), whereas an increase in permeability surface area product (mean, 5.9 ml/min/100 g) was associated with a shorter median progression-free survival time (4.7 months) and a shorter median overall survival time (10.6 months). An increase in permeability surface area product may indicate increased leakage from neoplastic vessels, which may be a precursor to the tumor invasion and metastasis. There were several limitations to our study. A study by Goh et al. [18] showed that a scan acquisition of 45 seconds was too short for reliable permeability measurement, and the 50 seconds in our study also was not long enough. The heterogeneity of our patient population in terms of staging and treatment can potentially affect our results. The size of the group in the follow-up study was small, and a larger sample size may be needed to verify our results. The manual drawing of regions of interest along the tumor margins is operator dependent and somewhat subjective. Future software applications may be useful for defining tumor margins with greater accuracy, reproducibility, and automation. Furthermore, analytic methods and commercial software packages are not directly interchangeable [20], which restricts clinical application of perfusion CT. We conclude that the perfusion parameters of lung cancer are reproducible. NSCLC with high perfusion is relatively sensitive to chemoradiation therapy. Perfusion CT is useful in predicting early tumor response and the prognosis of NSCLC after chemoradiation therapy. In the future, monitoring the response to therapy with perfusion CT may aid clinicians in customizing the treatment of each patient, improving patient selection for particular therapies, and avoiding nonproductive treatment regimens. AJR:193, October

7 Wang et al. References 1. 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: Werner-Wasik M, Xiao Y, Pequignot E, Curran WJ, Hauck W. Assessment of lung cancer response after nonoperative therapy: tumor diameter, bidimensional product, and volume a serial CT scan-based study. Int J Radiat Oncol Biol Phys 2001; 51: Cortes J, Rodriguez J, Diaz-Gonzalez JA, et al. Comparison of unidimensional and bidimensional measurements in metastatic non-small cell lung cancer. Br J Cancer 2002; 87: Watanabe H, Yamamoto S, Kunitoh H, et al. Tumor response to chemotherapy: the validity and reproducibility of RECIST guidelines in NSCLC patients. Cancer Sci 2003; 94: Erasmus JJ, Gladish GW, Broemeling L, et al. Interobserver and intraobserver variability in measurement of non-small-cell carcinoma lung lesions: implications for assessment of tumor response. J Clin Oncol 2003; 21: Grossi F, Belvedere O, Fasola G, et al. Tumor measurements on computed tomographic images of non-small cell lung cancer were similar among cancer professionals from different specialties. J Clin Epidemiol 2004; 57: Sekine I, Tamura T, Kunitoh H, et al. Progressive disease rate as a surrogate endpoint of phase II trials for non-small-cell lung cancer. Ann Oncol 1999; 10: Shepherd FA. Chemotherapy for advanced nonsmall-cell lung cancer: modest progress, many FOR YOUR INFORMATION choices. J Clin Oncol 2000; 18[suppl]:35S 38S 9. de Geus-Oei LF, van der Heijden HF, Visser EP, et al. Chemotherapy response evaluation with 18 F- FDG PET in patients with non-small cell lung cancer. J Nucl Med 2007; 48: Bruzzi JF, Munden RF. PET/CT imaging of lung cancer. J Thorac Imaging 2006; 21: Bellomi M, Petralia G, Sonzogni A, Zampino MG, Rocca A. CT perfusion for the monitoring of neoadjuvant chemotherapy and radiation therapy in rectal carcinoma: initial experience. Radiology 2007; 244: 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. Am J Neuroradiol 2006; 27: 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: Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: Kiessling F, Boese J, Corvinus C, et al. Perfusion CT in patients with advanced bronchial carcinomas: a novel chance for characterization and treatment monitoring? Eur Radiol 2004; 14: Ng QS, Goh V, Klotz E, et al. Quantitative assessment of lung cancer perfusion using MDCT: does measurement reproducibility improve with greater tumor volume coverage? AJR 2006; 187: Goh V, Halligan S, Hugill JA, Bassett P, Bartram CI. Quantitative assessment of colorectal cancer perfusion using MDCT: inter- and intraobserver agreement. AJR 2005; 185: Goh V, Halligan S, Hugill JA, Gartner L, Bartram CI. Quantitative colorectal cancer perfusion measurement using dynamic contrast-enhanced multidetector-row computed tomography: effect of acquisition time and implications for protocols. J Comput Assist Tomogr 2005; 29: Goh V, Halligan S, Hugill JA, Bartram CI. Quantitative assessment of tissue perfusion using MDCT: comparison of colorectal cancer and skeletal muscle measurement reproducibility. AJR 2006; 187: Goh V, Halligan S, Bartram CI. Quantitative tumor perfusion assessment with multidetector CT: are measurements from two commercial software packages interchangeable? Radiology 2007; 242: Sanelli PC, Nicola G, Tsiouris AJ, et al. Reproducibility of postprocessing of quantitative CT perfusion maps. AJR 2007; 188: Hermans R, Meijerink M, Van den Bogaert W, Rijnders A, Weltens C, Lambin P. Tumor perfusion rate determined noninvasively by dynamic computed tomography predicts outcome in headand-neck cancer after radiotherapy. Int J Radiat Oncol Biol Phys 2003; 57: Hermans R, Lambin P, Van der Goten A, et al. Tumoural perfusion as measured by dynamic computed tomography in head and neck carcinoma. Radiother Oncol 1999; 53: Garcia-Barros M, Paris F, Cordon-Cardo C, et al. Tumor response to radiotherapy regulated by endothelial cell apoptosis. Science 2003; 300: Ng QS, Goh V, Milner J, Padhani AR, Saunders MI, Hoskin PJ. Acute tumor vascular effects following fractionated radiotherapy in human lung cancer: in vivo whole tumor assessment using volumetric perfusion computed tomography. Int J Radiat Oncol Biol Phys 2007; 67: The comprehensive book based on the ARRS 2009 annual meeting categorical course on Ultrasound: Practical Sonography for the Radiologist is now available! For more information or to purchase a copy, see AJR:193, October 2009