HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Aberle, D. R. Articles by Sorensen, A. G. Search for Related Content PubMed PubMed Citation Articles by Aberle, D. R. Articles by Sorensen, A. G.

Imaging and Cancer: Research Strategy of the American College of Radiology Imaging Network¹

¹ From the Departments of Radiology of the University of California, Los Angeles (D.R.A.); Wake Forest University School of Medicine, Winston-Salem, NC (C.C.); the Brown Center for Statistical Sciences, Brown University, Providence, RI (C.G.); University of Virginia Health System, PO Box 800170, Charlottesville, VA 22908 (B.J.H.); Mayo Medical School, Rochester, Minn (C.D.J.); Yale University School of Medicine, New Haven, Conn (B.L.M.); Thomas Jefferson University School of Medicine, Philadelphia, Pa (D.G.M.); University of North Carolina School of Medicine, Chapel Hill, NC (E.D.P.); University of Pennsylvania School of Medicine, Philadelphia, Pa (M.D.S.); and Harvard University School of Medicine, Cambridge, Mass (A.G.S.). Received October 13, 2004; accepted October 27. Address correspondence to B.J.H. (e-mail: bjh8a@virginia.edu).

	ABSTRACT

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
The American College of Radiology Imaging Network (ACRIN) is a cooperative group funded by the National Cancer Institute and dedicated to developing and conducting clinical trials of diagnostic imaging and image-guided treatment technologies. ACRIN's six disease site committees are responsible for developing scientific strategies and resultant trials within the framework of ACRIN's five key hypotheses: (a) Screening and early detection with imaging can reduce cancer-specific mortality. (b) Less invasive image-guided therapeutic methods can reduce the mortality and morbidity associated with treating cancer. (c) Molecular-based physiologic and functional imaging can improve the diagnosis and staging of cancer, thus improving treatment. (d) Functional imaging can portray the effectiveness of treatment earlier and more accurately, thus reducing mortality and improving the likelihood of a cure. (e) Informatics and other "smart systems" can improve the evaluation of patients with cancer, thus leading to better and more effective treatments. This article details ACRIN's research strategy according to disease site through the year 2007.

© RSNA, 2005

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
The American College of Radiology Imaging Network (ACRIN) is a cooperative group funded by the National Cancer Institute formed to conduct clinical trials of diagnostic imaging and image-guided technologies. ACRIN has been conducting clinical trials since 1999. The organization has received funding for 2004–2007 to pursue 21 clinical trials on which it has been actively working. In addition, ACRIN has received supplemental funding to conduct screening trials of lung, breast, and colon cancer. Over 100 institutions have been qualified by ACRIN to participate in its trials.

The ACRIN trials address the major applications of imaging to cancer care: screening, diagnosis and staging, image-guided treatment, and measurement of response to treatment. To conduct its trials, ACRIN has developed collaborative relationships with other important elements of the cancer research community—most notably, the therapy cooperative groups, Specialized Programs of Research Excellence, industry groups, foundations, and patient advocacy groups. In some cases, these alliances are essential for subject recruitment, such as trials for which ACRIN must accrue patients at the point they present clinically, before they undergo imaging.

In addition to developing trials directed at specific cancers, ACRIN is iteratively developing an advanced informatics infrastructure to address a broad need to develop knowledge of the imaging characteristics of cancers as they relate to varying cancer genotypes and the behavior of neoplasms. All images from ACRIN trials are digitally archived at ACRIN's Philadelphia, Pa, headquarters. By collaborating with academic institutions and commercial entities, ACRIN will use this archive in the development and application of image analysis tools to allow automated image feature extraction and computer-aided detection and diagnosis.

Congruent with its new funding cycle, ACRIN has reorganized its scientific committee structure and constituencies. The scientific committees are of two types: modality committees and disease site committees (see the organizational chart on the ACRIN Web site, available at www.acrin.org, for details). The responsibility of the five modality committeesis to survey opportunities and to bring those opportunities to the disease site committees through the membership of modality committee constituents on disease site committees. This cross-fertilization is intended to be an efficient mechanism for the exchange of information concerning innovative technologies that will lead to the development of important clinical trials.

The six ACRIN disease site committees are responsible for the generation of scientific strategy for the organization within the context of ACRIN's overarching research strategy. Between 2004 and 2007, this overarching strategy will focus on the following five key hypotheses: (a) Screening and early detection with imaging can reduce cancer-specific mortality. (b) Less invasive image-guided therapeutic methods can reduce mortality and the morbidity associated with treating cancer. (c) Molecular-based physiologic and functional imaging can improve the diagnosis and staging of cancer and thus improve treatment. (d) Functional imaging can portray the effectiveness of treatment earlier and more accurately, thus reducing morbidity and improving the likelihood of a cure. (e) Informatics and other "smart systems" can improve the evaluation of cancer, leading to better and more effective treatments.

The individual disease site committee strategies make up the comprehensive ACRIN research agenda, which we believe also circumscribes the agenda for imaging and cancer. The pursuit of clinical trials based on these strategies is expected to provide information that will further ACRIN's principal goal of improving the length and quality of life in patients with cancer.

	Breast Cancer

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
The key clinical questions for breast cancer that should be addressed in imaging clinical trials are as follows: What methods can be used to improve the early diagnosis of breast cancer in special populations, particularly in young women and women with radiographically dense breasts? Can breast cancer screening be improved for these and other populations through adaptation of imaging diagnosis and treatment to patient risk and, if so, through the use of which imaging tools? What imaging tools can be used to help distinguish aggressive breast cancers from nonaggressive breast cancers, thus allowing therapies to be adapted to the tumor characteristics? Can we adapt therapy to tumor aggressiveness and use less aggressive (ie, percutaneous or transcutaneous) therapies for less aggressive malignancies? What imaging tools can be used to better assess the success of therapeutic regimens for breast cancer so that therapies can be adjusted when tumors are not responding?

	Rationale and How Imaging Might Be Used to Address These Questions

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
Adaptation of Diagnosis and Treatment to Patient Risk
The mainstay of breast cancer screening is still screen-film mammography. Its value as a test that helps to save lives by allowing early detection of breast cancer has been repeatedly demonstrated through seven screening trials in four countries over the past 40 years (1). Unfortunately, 40 000 women still die of breast cancer in the United States every year (2). The sensitivity of screening mammography is estimated to range from 83% to 95%, and the specificity is estimated to range from 90% to 98% (3). Sensitivity is lessened in women with dense breasts, with one author estimating it to be as low as 48% (4). As a result, conditions are not diagnosed with screening in many women with breast cancer; instead, conditions are diagnosed with palpation of a finding at physical examination or self-examination. In addition, there is a high probability of false-positive findings at screening in up to 10% of women (3), and it has been suggested that there is a 50% chance of an abnormal examination in women who undergo screening for 10 years (5).

The current mammographic screening guidelines in the United States are uniform across the entire population of women older than 40 years. Women at high risk for breast cancer might benefit from enhanced screening imaging tools. As an example, there is obvious need for improvement in breast cancer detection, both in increasing sensitivity in women with dense breasts and in increasing specificity. The ACRIN Digital Mammographic Imaging Screening Trial (ACRIN 6652), which is currently concluding patient follow-up and conducting reader studies, will enable researchers to determine whether digital mammography offers such an advance over traditional screen-film mammography. These concepts are also currently being tested in women at high risk for breast cancer in the ACRIN trials of screening breast magnetic resonance (MR) imaging (ACRIN 6667) and breast ultrasonography (US) (ACRIN 6666). Other potentially useful screening tools include genetic tests and serum proteomic markers, which might be combined with imaging screening to enhance accuracy and efficiency.

With the guidance of these trial results, more individualized screening tools and recommendations are potentially available within the next 5 years. Clearly, the development of those tools and guidelines should be a high priority for the ACRIN Breast Committee. Potentially useful technologies on the horizon that might be studied as part of ACRIN trials are digital mammography with tomography and three-dimensional reconstruction and dual-energy or digital subtraction mammography before and after the administration of intravenous contrast agents. These ACRIN trials will generate a vast archive of images that will be made available to academic and commercial entities for the development and testing of computer-aided detection and diagnosis.

Differentiation between More and Less Aggressive Breast Cancers
Ductal carcinoma in situ (DCIS) is not a cause of death per se. Breast cancer can only cause death in a patient after it has invaded and spread to other organs. DCIS is a substantial risk factor for the development of invasive disease, however, and its presence may usually precede the development of invasive tumor. However, the presence of DCIS does not indicate that an invasive tumor will inevitably develop (6). The diagnosis of DCIS has increased approximately tenfold in the United States and other developed countries since the introduction of population-based screening mammography. This suggests that some cancers that would have never injured a patient are being diagnosed and treated, thus causing unneeded morbidity and mortality (7).

DCIS is currently treated with either mastectomy or breast-conserving surgery and accompanied by radiation therapy and hormone therapy with tamoxifen. The recurrence rate is approximately 30% without radiation therapy and approximately 15% with it; there is no difference in mortality between the two therapies when compared with each other and when compared with mastectomy (2%). Recurrences tend to be 50% in patients with DCIS and 50% in patients with invasive breast cancers. As a result, some have argued that DCIS is not overtreated, that the increasing incidence of DCIS reflects more careful screening of the population, and that eventually this will yield lower breast cancer mortalities in the screened populations. Further evidence to support this viewpoint is found in the form of higher than expected numbers of invasive breast cancers in women with DCIS, increased risk of DCIS and invasive breast cancer in women treated for DCIS, and the results of genetic studies in women with DCIS and invasive breast cancer (8).

Ideally, however, there would be some mechanism for clinicians to determine which DCIS lesions are likely to lead to or become associated with invasive breast cancer and the death of the patient so that the more aggressive tumors can be treated aggressively and those that are not likely to kill the patient can be left alone or treated less aggressively. In general, high-grade DCIS is more likely to be associated with an invasive tumor. While there are some imaging features that loosely correlate with the grade of DCIS that is subsequently diagnosed, these findings are not very sensitive or specific. It is important for breast imagers to evaluate molecular imaging tests that, together with serum markers, might allow the aggressiveness of mammographically apparent lesions to be determined in vivo, perhaps precluding biopsy and even therapy for more indolent cases. One such potential tool is breast MR spectroscopy for tumor choline levels.

In addition, there are multiple tools that might be useful for the less invasive treatment of small early breast cancers, possibly including DCIS, should they be determined to be effective. These currently include radiofrequency (RF) ablation, microwave phased-array thermotherapy, high-intensity focused US, cryoablation, and laser therapy. The role of imaging in administration of the therapies and caring for the patient after therapy has been completed is yet to be determined. ACRIN trials will corroborate the effectiveness of less invasive image-guided therapies with current standard therapies to elucidate which women might not need open surgery.

Evaluation of the Effectiveness of Treatment
In ACRIN's trial of dynamic contrast agent–enhanced MR imaging as an intermediate marker for therapeutic effectiveness (ACRIN 6657), researchers will evaluate a possible role for breast MR imaging in determining whether novel chemotherapeutic regimens are working well in individual patients, particularly in the treatment of advanced breast cancer (stage 3 or higher). Such patients are at high risk of dying. If dynamic contrast-enhanced MR imaging proves successful in enabling physicians to predict which patients are responding to treatment, the effect on patients will be profound, as this technique will enable the early stoppage of ineffective treatment and the more rapid initiation of alternative therapy.

The use of MR imaging as a tool to monitor the efficacy of chemotherapy is really in its infancy. Patients who respond to chemotherapy may have typical contrast enhancement patterns over time, and the diagnostic criteria for interpreting MR imaging findings in the setting of chemotherapy may differ substantially from that used prior to chemotherapy. Other imaging tools that might be useful in the evaluation of the effects of chemotherapy over time in patients at high risk of death from breast cancer if therapy is unsuccessful—and that might be used in ACRIN trials—are MR spectroscopy, positron emission tomography (PET) with novel agents, US, and digital subtraction mammography.

	Gastrointestinal Cancers

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
The Gastrointestinal Committee chose the following as its research focuses: (a) methods to screen for colorectal polyps and cancer, (b) development of computer-aided detection tools for colorectal polyp and cancer detection, (c) assessment of patients with chronic liver disease to discriminate dysplastic regenerative nodules from hepatocellular carcinoma, (d) development of imaging tools or techniques to improve discrimination between benign serous pancreatic tumors from the premalignant or malignant mucinous tumors, (e) development of better methods to gauge response to therapy in patients with liver metastases, (f) banking of tissue and serum to correlate biomarker findings with imaging studies, and (g) percutaneous ablation of cancer for local control and reduced mortality. The Gastrointestinal Committee has the intent of developing clinical trials around these focuses for liver metastases and colorectal, hepatocellular, and pancreatic carcinomas.

	Rationales and Proposed Imaging Trials

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
Colorectal Cancer
Colorectal cancer causes substantial morbidity and mortality, especially in industrialized nations. It is the third most common cancer and second leading cause of malignant death in the United States, with an estimated 134 000 new colorectal cancer cases and 55 000 colorectal cancer deaths each year (2). As the natural history of colorectal cancer permits the recognition and curative treatment of both precursor adenomas and localized cancers, there is an enormous opportunity to save lives with early detection programs broadly applied to a general population. However, the potential efficacy and practicality of such a screening effort are compromised by limitations in the performance, comfort, and expense of available screening tests. Better tools are needed to more effectively screen for colorectal neoplasia.

Computed tomographic (CT) colonography (virtual colonoscopy), a noninvasive technique requiring only bowel preparation, is a structural examination of the entire colorectum that uses volumetric data acquired from a CT scanner combined with advanced computer software for image display. Controversy exists as to its sensitivity and specificity in the detection of large adenomas (especially in a screening population), as some authors have reported performance comparable to that of conventional colonoscopy (9), while others have reported only moderate sensitivity (10,11). A strong need exists to clinically validate widespread use of CT colonography in a screening population for the detection of colorectal neoplasia. In addition, the cost-effectiveness implications of observed performance outcomes need to be addressed. These needs are accommodated by ACRIN 6664—which serves to evaluate CT colonography in a screening population at high risk for colon cancer. Accrual of subjects began in February 2005.

In addition, all data from the trial will be archived and made available for academic and commercial concerns to test computer-aided detection programs aimed at improving observer performance and reducing variability.

Hepatocellular Carcinoma
Hepatocellular carcinoma is the most common primary hepatic tumor and one of the most common malignancies world wide. This devastating disease accounts for 17 300 new cases of cancer and 14 400 deaths yearly (2). It is believed that regenerative nodules in the liver, which develop as a result of cirrhosis, undergo transformation over the course of many years into dysplastic nodules and, ultimately, carcinoma. During the transformation between a dysplastic nodule and carcinoma, there are changes in the blood supply within these nodules. As the benign nodule changes into a malignancy, the major blood supply also changes from portal venous to arterial (12–14).

The natural history of this tumor leads many investigators to believe that there is high potential to use imaging to screen, detect, and treat many of these premalignant lesions or early cancers before metastases have developed. Imaging techniques that hold promise include MR imaging, CT, and US—especially with the use of novel intravenous contrast agents—for assessment of lesion perfusion. Radionuclide scanning with innovative targeted contrast agents holds the promise of indicating the metabolic or physiologic nature of abnormalities, which may allow better differentiation of benign hyperplastic nodules from malignancy. ACRIN trials that involve obtaining tissue specimens will allow correlation of imaging findings with molecular expression. This may allow identification of hepatocellular carcinoma subtypes, thus opening the door to image-driven customized therapeutic approaches.

There is also a role for ACRIN in the investigation of less invasive treatment for hepatocellular carcinoma. Treatments of hepatocellular carcinoma with demonstrated efficacy include surgical resection and liver transplantation if lesions are small in number and size. For patients who do not fulfill criteria for resection or transplantation, however, treatment options are poor. RF ablation provides a potentially useful treatment option that may be superior to other available nonsurgical treatments (15,16). In addition to the possibility of prolonging the survival of patients who are not surgical candidates, RF ablation can be used in transplantation candidates to treat small hepatocellular carcinomas that arise while a patient is awaiting transplantation.

Cystadenocarcinoma of the Pancreas
Mucous cystadenocarcinomas are not reliably differentiated from benign serous tumors by their morphologic appearance (17,18). As a result, patients with asymptomatic benign neoplasms may undergo unnecessary surgery. However, serous tumors contain abundant low-viscosity glycogen-rich fluid, while mucinous tumors contain high-viscosity mucinous fluid. ACRIN trials of imaging methods that could be used to distinguish these tumors according to their fluid differences might prove to be effective diagnostic tools. New imaging methods, such as diffusion MR imaging and MR spectroscopy, hold promise in this regard and are ripe for investigation.

Assessing the Therapeutic Response of Liver Metastases
Liver metastases are the most common malignant neoplasm to affect the liver. Despite great advances in imaging technology that allow physicians to directly visualize liver metastases and increase the ability of physicians to detect these lesions with ever-increasing precision, our ability to measure the effect of treatment on these lesions remains limited.

The Gastrointestinal Committee proposes trials in which researchers will address the need to better assess the volume of residual tumor in the liver (19,20) and the functionality of the remaining tumor after therapy (21,22). New CT and MR imaging technology allows acquisition of three-dimensional volumes of the liver, which may represent an improvement on the traditionally used Response Evaluation Criteria in Solid Tumors. Functional assessment of tumors is also possible with assessment of contrast enhancement within lesions and with the use of PET scanning. Potentially better ways to determine viable tumor within detectable lesions include the use of targeted molecular agents, perhaps with fusion technologies, such as combined PET/CT. In addition, the committee proposes that ACRIN collaborate with therapeutic cooperative groups to test the potential of new molecular imaging technologies to serve as intermediate markers of the effectiveness of treatment, particularly with regard to predicting pathologic response and extended survival.

Surgical resection is the only treatment for colorectal metastases that has been shown to improve 5-year survival. As with hepatocellular carcinoma, treatment options for patients who are not candidates for surgery are poor. RF ablation is a treatment that could perhaps extend survival and potentially improve quality of life by improving local control. The Gastrointestinal Committee recommends a trial that would serve to test this hypothesis. Such a trial could also serve to evaluate specific characteristics of lesions, such as size, location, and blood supply, which are used to predict the success or failure of RF ablation.

	Cancer in the Thorax

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
Although predominantly focusing on lung cancer, the Thoracic Disease Committee also will consider trials of cancer arising from extrapulmonary sites. Its strategy through 2007 is to focus on five major activities. The first of these is continuation of the current National Lung Screening Trial (ACRIN 6664), including analysis of early data.

The second activity is measurement of tumor response to therapy in patients with lung cancer. Specifically, researchers want to answer the following questions: (a) Does measurement of tumor volume on CT scans provide an earlier and more reliable indication of response to antitumor therapy than uni- or bidimensional measurement? (b) Is measurement of the standardized uptake value (SUV) or the volume-corrected SUV on PET images a better indicator of tumor response than measurements of tumor size on CT scans?

The third activity is assessing the single pulmonary nodule for malignancy with PET. Specific questions include (a) Is 3-deoxy-3-flourine 18-fluorothymidine (FLT) PET superior to 2-deoxy-2-flourine 18-fluoro-D-glucose (FDG) PET in the distinction of benign from malignant pulmonary nodules? (b) Are size-corrected SUVs more accurate than SUVs that are not size corrected in the distinction of benign from malignant pulmonary nodules?

The fourth activity is assessing the role of FDG PET in determining treatment of esophageal cancer—including (a) in patients receiving therapy for localized esophageal cancer, can PET be used to distinguish those who are responding to therapy from those who are not responding? (b) Does incorporation of PET in the follow-up and treatment planning of these patients improve survival?

The fifth activity is RF ablation of lung tumors. Specifically, in patients with medically inoperable early stage lung cancer, can RF ablation, in conjunction with either stereotactic radiation therapy or catheter-delivered brachytherapy, be used to improve survival compared with stereotactic radiation therapy?

	Rationales and Imaging Proposals

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
National Lung Screening Trial
The National Lung Screening Trial is a collaboration of ACRIN and the Lung Screening Study of the National Cancer Institute Prostate, Lung, Colon, and Ovary trial; it is intended to determine if CT screening for lung cancer reduces lung cancer–specific mortality in high-risk individuals. The National Lung Screening Trial began enrollment in September 2002 and concluded enrollment in February 2004, with a total of 53 418 participants (18 893 patients enrolled in ACRIN, and 34 525 enrolled in the Lung Screening Study). Screening will continue through the middle of 2006, and health information will be obtained through 2009. Early publications will include results of prevalence and incidence screening, results of the recruitment process, development of image quality control standards, and the influence of the screening process on smoking habits and beliefs.

Measurement of Therapeutic Response in Patients with Lung Cancer
Lung cancer is the most common cause of cancer death in both men and women in the United States and worldwide. Evaluating response to the different treatment regimens within clinical trials for lung cancer and comparing these results with other trials requires consistent criteria for determining response. In 1994, the World Health Organization criteria for reporting the results of cancer treatment were reviewed, and revised guidelines known as Response Evaluation Criteria in Solid Tumors were proposed, with treatment response determined by using a single measurement of the largest tumor diameter in the transverse plane. Although interobserver agreement in size determination of small pulmonary tumors is good when spiral CT is used to assess well-defined tumors, considerable variability exists in measurements of lung tumors performed by different readers if the lesion is irregular (23). These interobserver differences in measurements can erroneously affect the data and outcomes of clinical trials. Adaptation of tumor measurements by using volumetric analysis tools may allow an earlier and less variable assessment of tumor response (24).

The development of novel agents, such as gene therapy, as well as treatment protocols that target tumor biology—including tumor cell proliferation and invasion, angiogenesis, and metastasis—further complicate the measurement of tumor response. Antitumor effect in many of these regimens is cytostatic and, unlike anticancer cytotoxic agents, may not cause regression in tumor size. Anatomic measurements consequently may not adequately reflect the efficacy of cytostatic agents in cancer trials. Molecular imaging technologies, such as PET, can enable analysis of the metabolic activity of tumors and will be the subject of ACRIN trials to assess their capabilities to provide early and accurate information on the effectiveness of these new therapeutic agents. This will be performed by embedding ACRIN imaging trials in treatment trials of novel therapeutic agents. Researchers will perform serial measurements of tumor size on CT scans, including unidimensional, bidimensional, and volume measurement, and PET SUV measurements of the primary tumor sites before and after treatment. Researchers will then analyze the results in the context of pathologic response of the tumor and patient survival. ACRIN also will focus on optimal ways of using PET data to assess treatment effect, as PET measurements can be subject to error. The committee proposes to test the possibility that volumetric analysis of SUV values, in a manner similar to anatomic volumetric analysis, may be more reliable in the prediction of tumor response.

PET Assessment of the Solitary Pulmonary Nodule
With the implementation of screening CT for lung cancer and the frequent detection of pulmonary nodules with CT, a noninvasive means of detecting neoplasia is necessary. FDG PET is more sensitive (sensitivity of 92%–96%) and specific (specificity of 78%–96%) than CT in the detection of malignancy in pulmonary nodules that are larger than 1 cm in diameter. Despite its improved depiction of neoplasia, there are tumors that are not consistently detected with FDG PET. Tumors such as carcinoid bronchioloalveolar cell carcinoma and other well-differentiated adenocarcinomas are frequently determined to be false-negative with FDG PET. FDG is also taken up by inflammatory and infectious processes that can mimic neoplasia. Recent studies have shown that FLT is a useful biomarker for tumor cell proliferation (25). Preliminary studies have shown that FLT PET may be better than FDG PET in distinguishing benign nodules.

The committee proposes a multicenter study to evaluate the sensitivity and specificity of FDG PET and FLT PET in the detection of malignancy of pulmonary nodules on the basis of estimation of glucose metabolism with FDG PET and thymidine turnover rate with FLT PET. This study could allow researchers to determine if the uptake of FDG and FLT varies according to tumor characteristics, such as specific tumor types (adenocarcinoma—including bronchioloalveolar carcinoma type, large cell, squamous cell, carcinoid), tumor grade (high vs low grade), and differentiation (good vs poor).

PET Evaluation of Treatment for Esophageal Cancer
Esophageal cancer is the sixth leading cause of death from cancer worldwide. Patients with localized esophageal cancer are treated with surgical resection. Trials have also been conducted on preoperative radiation therapy, preoperative chemotherapy, and combined preoperative chemoradiation. There may be a subset of patients with local-regional esophageal cancer who experience a substantial pathologic response to multimodality therapy. If patients who respond to treatment could be distinguished from those who do not respond to treatment, aggressive treatment could be pursued in the first group of patients and discontinued in the second. In preliminary studies, FDG PET was used to predict disease-free survival and overall survival on the basis of change in the SUV at the primary site on pre- and posttreatment scans (26). The relative change in SUV has also been shown to correlate with the percentage of viable cells in the resected surgical specimen (27,28).

In patients with surgically resectable esophageal carcinoma, a prospective randomized trial incorporating pre- and posttreatment FDG PET scanning is proposed. PET data would be correlated with patient survival and morbidity as they relate to chemoradiation, surgery, and palliative care. Tissue specimens would be collected so that imaging phenotype could be correlated with cancer genotype.

RF Ablation of Lung Cancer
Surgical resection is the standard of care in patients with early-stage non-small cell lung cancer. Some patients with potentially curable lung cancers are not candidates for surgery because of limited pulmonary reserve, coexisting morbidity, or both. In these patients, external beam radiation therapy, with or without chemotherapy, is the most frequently used treatment regimen. The 5-year survival rate with this approach ranges from 10% to 30%. In this patient population, the noninvasive strategies being investigated are based on gaining better local control of the tumor and include RF ablation, brachytherapy, stereotactic radiation therapy, and cryotherapy (29–31).

Because RF ablation targets the center of the tumor, combined therapy with external beam radiation therapy—which has maximal effect on the periphery of the tumor—may prove to be advantageous, particularly in larger tumors. Combined RF ablation and catheter-delivered brachytherapy with either high-activity iridium 192 seeds (temporary implants) or low-activity iodine 125 seeds (permanent implants) has also been reported (32). The appropriate follow-up of patients treated with local therapy has not been established.

The committee proposes a multicenter, randomized controlled trial for the purpose of evaluating outcomes in nonsurgical patients with stage I non-small cell lung cancer. In this study, researchers would compare RF ablation in combination with stereotactic radiation therapy and RF ablation in combination with brachytherapy to stereotactic radiation therapy alone. Follow-up would be performed with a combination of contrast-enhanced CT and FDG PET. This trial could be used to evaluate the efficacy of these approaches for local control of lung cancer and develop standards for follow-up and further treatment.

	Gynecologic Malignancies

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
In each woman with a gynecologic malignancy, imaging should facilitate the choice of treatments that will best cure or delay the progression of malignant disease so that adverse effects of unnecessary surgery or other treatment can be avoided. The determined treatments should then be targeted to the tumor, thus sparing adjacent organs and tissues that might be damaged by the treatment. The ACRIN Gynecologic Committee chose the following focuses to guide their strategy for research. First, the committee chose to focus on evaluation of women with cervical cancer to guide surgical and chemoradiation therapy decisions, including evaluation of local extent of disease, discernment of characteristics of the primary tumor, and assessment of lymph node metastases. Second, they chose to determine the need for surgical staging in women with endometrial carcinoma. Third, the committee chose to improve the selection of women with adnexal masses needing oophorectomy. Fourth, they chose to improve determination of which women with recurrent ovarian carcinoma would benefit from optimal secondary cytoreduction surgery and prediction of early response to chemotherapy for recurrence.

	Rationales and Imaging Studies

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
Cervical Cancer
Patients with carcinoma confined to the cervix or upper vagina (stage IIA or lower) can be treated with either chemoradiation therapy or radical hysterectomy with bilateral pelvic lymphadenectomy, whereas those with carcinoma with parametrial or lymphatic involvement (stage IIB or higher) are now treated with chemoradiation therapy (33–35). Treatment failures are generally because of resistance to chemoradiation of the primary tumor, as well as occult lymphatic metastases that are not resected or irradiated. Primary chemoradiation without surgery would generally be preferred if the nodal or parametrial tumor or the potential for involved margins at surgery could be accurately identified before surgery (34). More accurate local staging would also allow refining of the volume of tissue treated with targeted radiation therapy, in contrast to the current practice of irradiating the entire pelvis to ensure treatment of undetected metastases in parametrial tissues and pelvic lymph nodes. Data from a collaborative trial between ACRIN and the Gynecologic Oncology Group, or GOG, (ACRIN 6651/GOG 183) are currently being analyzed and used to compare the staging accuracies of CT, MR imaging, and clinical International Federation of Gynecology and Obstetrics standards in patients with stages IB and IIA cervical carcinoma. Accuracy of CT and MR imaging for anatomic delineation should be reevaluated in a few years, as both of these techniques continue to improve.

Angiogenesis and tumor hypoxia are crucial factors for tumor development and progression. The inhibition of angiogenesis is considered one of the most promising strategies that may lead to novel cancer therapies (36). A noninvasive in vivo measure of angiogenesis could enhance the value of studies involving these new treatment strategies (37). Tumor hypoxia contributes to treatment failure (38). Identification and measurement of tumor hypoxia might help physicians predict outcome and select patients for therapies designed to increase the radiosensitivity and chemosensitivity of hypoxic tumor cells (39).

Doppler US signals can be augmented after intravenous injection of microbubble contrast agents, thus improving the correlation between Doppler US and parameters of angiogenesis. Perfusion imaging with either CT or MR imaging techniques can be used to depict both angiogenesis and hypoxia in tumors (36,40). Perfusion CT or MR imaging can be potentially accomplished as part of a comprehensive examination that is also used to evaluate the local extent of the tumor and detect lymphadenopathy. Perfusion imaging with CT and MR imaging and perhaps with advanced US techniques that use novel contrast agents can be correlated in a multicenter trial with parameters such as pathologic and patient response to chemoradiation therapy.

Since metastases are often present within lymph nodes that are not enlarged, decisions regarding extent of radiation therapy have been dependant on extraperitoneal surgical staging. However, preirradiation surgical staging has been associated with an increase in late radiation-related morbidity.

FDG PET is used to detect lymphatic metastases because of their abnormally high metabolic rate. The accuracy of PET is improved by fusing PET and CT images obtained during the same examination. Most investigators are convinced that PET/CT is more accurate in the diagnosis of lymphatic metastases than are current CT or MR imaging techniques (41–43); however, to our knowledge, its negative predictive value has not yet been determined in a large surgically validated series. MR imaging after intravenous administration of lymphotropic ultrasmall superparamagnetic iron oxide nanoparticles can also be used to distinguish between tumor and benign tissue in normal-sized lymph nodes (44). There has, however, been no published trial on the accuracy of this technique and no comparison with PET/CT in patients with cervical carcinoma. ACRIN and the Gynecologic Oncology Group are preparing a protocol to evaluate PET/CT and ultrasmall superparamagnetic iron oxide–enhanced MR imaging for the diagnosis of lymphatic metastases in patients with locally advanced cervical carcinoma.

Endometrial Cancer
Endometrial cancer is highly treatable; 73% of affected women have disease localized to the uterus, with an estimated 5-year relative survival rate of 96%. Many clinicians omit surgical staging in women thought to be at low risk for extracorporeal extension because of the potential morbidity of surgical staging and because most patients will not have metastatic spread. Women with unfavorable risk factors that predict extrauterine metastasis, such as myometrial invasion, tumor grade, histologic characteristics, tumor size, and lymphovascular space invasion, are selected for surgical staging with pelvic and paraaortic lymph node sampling. In patients with lymphatic metastases, maximal cytoreduction has correlated strongly with duration of survival (45).

The potential benefits of an accurate preoperative estimation of the extent of extrauterine disease include a tailored surgical approach and streamlined triage of patients to appropriate clinical personnel for therapy (eg, gynecologic oncologist vs general gynecologist). The most important predictive measure for preoperative staging with imaging is the negative predictive value (so that the staging requirement is not inappropriately deferred). US and MR imaging have been more effective than CT in the diagnosis of deep myometrial invasion, but their negative predictive values have been less than 90%. Evaluation of tumoral blood flow with Doppler US has shown initial promise in the identification of tumor aggressiveness. The committee recommends that perfusion imaging and contrast-enhanced vaginal Doppler US be compared with surgical staging and findings of a pathologic examination. In addition, ACRIN should explore the value of PET/CT and ultrasmall superparamagnetic iron oxide–enhanced MR imaging to determine if either of these tests should be used routinely for preoperative diagnosis.

Adnexal Masses
In most large series, about two-thirds of adnexal masses removed for suspected malignancy are in fact benign (46–48). In many of these women, surgery could have been avoided if benign disease could have been diagnosed before surgery. Some reports suggest that complex masses can be characterized by using a combination of gray-scale US appearance and Doppler vascular morphology along with Doppler waveforms, as well as use of microbubble US contrast agents. Other series suggest that MR imaging is currently the most accurate imaging study for diagnosis of adnexal masses (49,50). FDG PET performed after an inconclusive US examination can improve the specificity for diagnosis of ovarian cancer, although it is less sensitive for borderline tumors. In designing trials to test these technologies, the key issues are negative predictive value (so that surgery is not delayed in women with early ovarian carcinoma), overall cost effect, and patient outcomes.

Ovarian Cancer
Approximately 80% of patients with epithelial ovarian cancer treated with maximal surgical cytoreduction followed with platinum and taxane-based primary therapy enter complete clinical remission. Unfortunately, 70% of optimally debulked patients experience a relapse at a median duration of approximately 18 months. Treatment options for patients with recurrent disease are dependent on time from prior treatment, previous response to chemotherapy, and extent of disease. Complete or optimal secondary surgical cytoreduction has been associated with prolonged survival in selected patients (51). However, no survival benefit has been shown for suboptimal secondary cytoreduction. Thus, the objective of imaging in the detection of recurrent ovarian carcinoma is to identify patients who are likely to achieve a complete or at least optimal secondary surgical cytoreduction.

Traditionally, we have classified patients with relapse at less than 6 months follow-up as platinum resistant and have selected nonplatinum agents for treatment; those with relapse at greater than 6 months are termed platinum sensitive and are administered a platinum compound or, more recently, a platinum-based combination treatment. Response assessment may require between two and four cycles of therapy, with associated side effects, until response manifests as either a decline in the serum CA-125 level or a radiographic or symptomatic improvement. Patients are often treated empirically with multiple sequential agents, with intervals separated by the time required to assess response.

FDG PET/CT has shown improved accuracy in the identification of bulky recurrent ovarian carcinoma, and it has been used to show early response to chemotherapy before changes in size are visible on conventional images (52). PET/CT and surgical results should be correlated, with overall survival and quality of life as the primary end points in women whose test results are positive for CA-125 but have a negative or equivocal CT scan, as well as in women with a recurrence that appears to be localized on CT scans. In addition, ACRIN should test the hypothesis that a PET/CT response after one cycle of therapy might potentially serve as an in vivo predictor of subsequent clinical response. The magnitude of PET/CT response should be evaluated as a possible predictor of the duration of chemotherapy response. This could be addressed in conjunction with any phase II study of a cytotoxic agent that requires depiction of measurable disease. Changes in PET/CT after one cycle of therapy would be correlated with subsequent clinical response rate and time to treatment failure.

	Genitourinary Cancers

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
The Genitourinary Committee will focus on the following issues: (a) evaluation of new imaging methods to enable preoperative staging of cancer as a means of improving the choice of treatment; (b) assessment of new methods of improving surveillance for tumor recurrence, including the use of smart systems to improve tumor recognition and further understanding of the natural history of these cancers; and (c) establishment of percutaneous methods of treating genitourinary malignancies in a less invasive manner.

Specifically, the committee will develop trials related to these considerations for the common genitourinary tract tumors of the prostate, bladder, and kidney.

	Rationales and Imaging Approach

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
Prostate Cancer
In men, prostate cancer is the second leading cause of death and the most common cancer diagnosed. Screening for prostate cancer with prostate-specific antigen is well established; however, compliance is variable, and the appropriateness of such screening is controversial (53). Radical prostatectomy and radiation therapy remain the established definitive treatment options. Reliable localization of prostate cancer is of increasing therapeutic importance as a result of the emergence of disease-targeted ablative therapies, such as interstitial brachytherapy, intensity-modulated radiation therapy, high-intensity focused US, and cryosurgery (54). Such focal treatment holds the promise of substantially reducing the morbidity associated with treatment of the entire prostate, whether with surgery or with radiation.

The selection of appropriate treatment options and the use of focal therapy all require accurate noninvasive evaluation of disease extent and aggressiveness. Unfortunately, all current methods of assessment are of limited accuracy. MR imaging and spectroscopy have shown substantial promise as innovative radiologic tools for use in the local evaluation of prostate cancer (55–57). Early data suggest that these modalities can positively affect treatment planning (58,59). A currently open ACRIN trial (ACRIN 6659) will enable researchers to answer key questions about the utility of these technologies and help guide future definitive therapies. This trial will enable researchers to evaluate the contribution of MR spectroscopy (eg, color mapping of MR spectroscopic information) to MR imaging in the localization and determination of the extent of prostate cancer.

Bladder Cancer
Transitional cell carcinoma of the bladder treated with radical cystectomy with limited or extended pelvic lymph node dissection has been studied since before 1946 (60). The involvement of local-regional lymph nodes varies and increases with extent of tumor invasion. Despite the use of regional lymphadenectomy at the time of radical cystectomy, there are few data from randomized trials to define an association between the extent of surgical lymph node resection and recurrence or survival.

Recent evidence suggests that lymph node imaging with MR and iron-based contrast agents may improve preoperative diagnosis and help better direct surgical therapy (44). Thus, selective limited lymph node dissections may be more precisely performed. The committee proposes a collaborative trial appended to an American College of Surgeons Oncology Group, or ACOSOC, trial (ACOSOG Z7031), which is a prospective evaluation of the benefit of a limited versus an extended pelvic lymphadenectomy performed at radical cystectomy for bladder cancer. The imaging arm would use a novel contrast agent, ultrasmall superparamagnetic iron oxide (Combidex; Advanced Magnetics, Cambridge, Mass), to image the internal architecture of lymph nodes before surgery and enable researchers to determine whether this approach improves the predictive value for determining the extent of lymphadenectomy. PET and other molecular imaging technologies may also prove useful for tumor staging and lymph node evaluation, and they may be the focus of additional ACRIN trials.

A second focus related to transitional cell carcinoma of the urinary tract will be surveillance of patients who have shown a propensity to develop urothelial cancers. The biology of urothelial cancers is considered a "field theory," in that once a cancer has been discovered, all the remaining urothelium remains at risk for the development of cancer, and lifelong surveillance becomes a necessity. ACRIN will evaluate screening and surveillance imaging techniques, such as x-ray CT urography for patients with urothelial carcinoma of the bladder and other urothelial neoplasms. ACRIN will use the image archive generated with such trials to test the value of sophisticated three-dimensional software in the manipulation of images as a means of providing enhanced diagnostic information.

Renal Cell Carcinoma
Most renal cell carcinomas are discovered today by means of serendipity due to the increased use of CT, US, and MR imaging performed for nonrenal causes. As a result, more renal cell carcinomas are being discovered at a smaller size and, likely, a lower stage. Cross-sectional imaging with CT or contrast-enhanced MR imaging offers a fast and accurate method to closely follow and observe the behavior of small solid renal neoplasms. Newer volumetric software and three-dimensional reconstruction promise improvement over conventional transverse imaging. Functional (molecular) imaging techniques with PET and radiopharmaceauticals such as ¹¹C-acetate performed to enable lesion detection and characterization may have value in following small renal cell carcinomas, as increased metabolism may occur before changes in the size of a neoplasm. The appropriate use of these imaging technologies to better characterize and follow incidentally discovered renal masses will require well-controlled clinical trials.

Definitive treatment for renal cell carcinoma remains partial or total nephrectomy with open or laparoscopic techniques. However, new therapies, such as RF ablation, have generated much interest in minimally invasive image-guided approaches. RF ablation for small lesions seems promising, but follow-up of patients treated with RF ablation proves challenging, since we have yet to learn the precise posttreatment patterns, as depicted on cross-sectional images. ACRIN trials of RF ablation and other less invasive image-guided treatment technologies could elucidate the answers to such questions.

Informatics
In addition to its organ-specific trial activities, the Genitourinary Committee proposes a cross-cutting effort to develop genitourinary radiology ontology, which might build on current activities of the National Library of Medicine Unified Medical Language System project. This project has three components: (a) metathesaurus, which contains information about biomedical concepts and terms from controlled vocabularies and classifications used in patient records, administrative health data, and bibliographic and full text databases; (b) SPECIALIST Natural Language Processing System, which is a lexicon of biomedical terms (available at: www.specialist.nlm.nih.gov); and (c) semantic networks, which categorize concepts represented in the metathesaurus with links between the concepts, thus providing a structure for the representation of important relationships in the biomedical domain. Building such "smart databases" for genitourinary diseases could be an important step into the future of data sharing beyond merely the transfer of clinical images. Such an effort would also lead to collaboration with those who have developed artificial neural networks in genitourinary cancer (61,62).

	Malignancies of the Head and Neck and Central Nervous System

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
A number of malignancies affect the head and neck and the central nervous system. The committee’s strategic focus initially is on primary cerebral gliomas and squamous cell carcinoma in the head and neck. These are the most common brain and head and neck malignancies, respectively, and they are associated with substantial morbidity and mortality. The prevalence of these tumors and their dismal prognoses provide substantial motivation for research.

With regard to cerebral gliomas and squamous cell carcinomas, the committee will develop clinical trials in which researchers will (a) evaluate imaging technologies that may allow for less invasive differentiation of malignant tissue from other abnormalities, such as hemorrhage or fibrosis; (b) use metabolic imaging, physiologic imaging, or both to predict the clinical outcomes of patients with brain tumors; (c) investigate the potential of imaging technologies to improve the staging of head and neck cancers to improve outcomes; and (d) evaluate the potential of imaging technologies to aid in the prediction of the effectiveness of treatment.

	Rationales for Investigation

TOP ABSTRACT INTRODUCTION Breast Cancer Rationale and How Imaging... Gastrointestinal Cancers Rationales and Proposed Imaging... Cancer in the Thorax Rationales and Imaging Proposals... Gynecologic Malignancies Rationales and Imaging Studies... Genitourinary Cancers Rationales and Imaging Approach... Malignancies of the Head... Rationales for Investigation REFERENCES

 
Primary Cerebral Gliomas
There are substantial differences between high-grade cerebral gliomas and neoplasms in other body locations, such as the colon, breast, and lung. Cerebral glioma is relatively uncommon, with an estimated 18 400 new primary intracranial gliomas and 12 600 deaths in the United States in 2004 (63). While its low prevalence precludes consideration of screening for cerebral glioma from a cost perspective, detection is rarely difficult. Conventional MR imaging techniques such as T2-weighted or fluid-attenuated inversion-recovery images and contrast-enhanced MR imaging are very sensitive for detection of parenchymal lesions. On the other hand, characterization of a lesion as a neoplasm rather than an infectious or inflammatory process can be problematic. Another key difference from other types of cancer is that patients with high-grade gliomas have 5-year survival rates of less than 15% (64), and high-grade gliomas progress rapidly, which reflects the lack of effective treatments. Thus, cerebral gliomas cause substantial morbidity and mortality.

For these reasons, our proposed imaging research strategy is to focus on imaging technologies that permit rapid and accurate characterization of a brain lesion as a neoplastic process (differential diagnosis) and that accurately reflect tumor viability, activity, or both (treatment monitoring and planning), with a particular eye toward the use of these techniques to aid in the development of effective therapies. Recently developed MR and nuclear imaging tools have become available that have great potential to enable researchers to address these two issues; therefore, there is a high likelihood that patient care may change and disease-associated morbidity and mortality may be potentially reduced.

A common difficult clinical scenario is the task of distinguishing neoplasm from infection, inflammation, or infarction or vascular causes with conventional imaging (65). It is imperative not to delay therapy by awaiting follow-up data; conversely, one does not want to perform unnecessary biopsy or surgery. Because of the rapid progression of high-grade gliomas, early accurate characterization is crucial for instituting appropriate therapy quickly, which may increase survival time. Furthermore, the therapeutic approach for a low-grade lesion may be different from that for a high-grade lesion. Hence, accurate diagnosis, including assigning a tumor grade, would be of great clinical value.

Our second focus in the evaluation of early therapeutic response relies on accurate and early assessment of tumor viability, activity, or both. Such assessments are imperative for quick adaptation of treatment regimens in patients for whom life expectancy is currently very limited and changes to alternative therapies on a rapid basis is crucial.

In addition, the noninvasive assessment of tumor viability, activity, or both could greatly aid the development and testing of new therapies. Imaging techniques that provide "biomarkers" of tumor pathobiology, such as angiogenesis, vascular permeability, and metabolic activity, might permit rapid noninvasive evaluation of the efficacy of many experimental therapies before more conventional morphologic (eg, tumor size, vasogenic edema) and clinical (eg, neurologic scores, death) parameters would be expected to change. For example, high-grade gliomas appear to require an increased blood supply compared with normal brain tissue, and there is a strong correlation between increased tumor neovascularity (eg, irregular enlarged microvessels recruited via angiogenesis) and malignancy (66). Drugs that prevent angiogenesis and consequently inhibit tumor growth are actively being pursued by researchers because of the lack of success of standard cancer treatments. To evaluate the clinical potential of these new drugs, a standardized widespread method for early noninvasive monitoring of their efficacy would be tremendously beneficial. The recovery of tissue for genomic and proteinomic assessment, so as to best develop molecular imaging tools in the future, will be an important feature of the committee’s trial portfolio because of the potential of molecular diagnoses in the future.

Squamous Cell Carcinoma in the Neck
There were an estimated 28 260 new cancers and 7230 deaths from lesions of the oral cavity and pharynx in the United States in 2004, with almost all of these being squamous cell carcinoma (63). Curative resection can be disfiguring and cause severe functional limitation; for this reason, the minimally required surgical extent is desired, and it is crucial that staging be accurate to remove nodal metastases but spare as much function as possible.

The presence or absence of nodal disease is one of the most important factors in predicting disease recurrence and survival in patients with squamous cell carcinoma; for example, the presence of nodal metastases can reduce 5-year survival by as much as 50% (67,68). Unfortunately, the surgical morbidity associated with extensive neck dissection is substantial, and some reports suggest that two-thirds of patients who do not have clinically palpable lymph spread and yet undergo surgical exploration to rule out microscopic disease end up not having nodal disease. They experience the morbidity of neck dissection without benefit. Similar concerns are present in monitoring patients after initial surgical therapy. Attempts to develop scintigraphic, dye-based, or other staging techniques are still controversial and largely confined to single-center studies. The lack of success of these approaches has been in part because the radiotracers or dyes used to date have not been cancer-specific markers. More specific markers, such as radiolabeled thymidine—which is a PET tracer—might provide more specific information about the status of lymph nodes and eventually reduce the need for surgical exploration while providing accurate nodal status (67).

Potential Trials
There are some commonalities in the approach with which the committee intends to address the two neoplasms. In addition, problems that can be solved with current technology should naturally come before those that require new technical solutions. Thus, our short-term priorities include designing multicenter trials to (a) determine the prognostic value of MR spectroscopy in newly identified intraaxial mass lesions (differential diagnosis), (b) determine the prognostic value of perfusion MR imaging in newly identified intraaxial mass lesions and the role of perfusion MR indexes as a surrogate for degree of tumor angiogenesis and a marker for response to therapy, (c) determine the utility of FLT in staging squamous cell carcinoma, (d) determine the role of diffusion-weighted or diffusion-tensor imaging as a marker of acute cytotoxicity, an