Utilization of Diagnostic Studies in the Pretreatment Evaluation of Invasive Cervical Cancer in the United States: Results of Intergroup Pro
http://www.100md.com
《临床肿瘤学》
the Department of Radiology, University of Miami Medical School, Miami, FL
Department of Radiology, Memorial Sloan-Kettering Cancer Center, New York, NY
Department of Radiology, Thomas Jefferson University Hospital, Philadelphia, PA
Center for Statistical Sciences, Brown University, Providence, RI
Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY
Department of Medical Oncology, Mayo Clinic, Rochester, MN
College of Medicine Division of Gynecologic Oncology, Lee Moffitt Cancer Center, Tampa, FL
ABSTRACT
PURPOSE: To review the current utilization of diagnostic tests prescribed by the International Federation of Gynecology and Obstetrics (FIGO) clinical staging guidelines in the pretreatment work-up of invasive cervical cancer, and to compare the data with those of previous patterns of care studies.
PATIENTS AND METHODS: This interdisciplinary American College of Radiology Imaging Network/Gynecologic Oncology Group prospective clinical trial was conducted between March 1, 2000, and November 11, 2002. Twenty-five participating institutions, all from the United States, enrolled a total of 208 patients. Only patients scheduled for surgery with biopsy-confirmed cervical cancer of clinical FIGO stage IB or higher were eligible. The patterns of care data analysis was based on 197 patients who met all inclusion criteria. The conventional FIGO-recommended tests used for pre-enrollment FIGO clinical stage classification were at the discretion of the treating physician; overall frequency of use was tabulated for each test.
RESULTS: Use of cystoscopy (8.1%) and sigmoidoscopy or proctoscopy (8.6%) was significantly lower than in 1988 to 1989 (P < .0001 in each instance). Intravenous urography was used in only 1% of patients as compared with 42% in 1988 to 1989 and 91% in 1983. No patient included in the data analysis had barium enema or lymphangiography. Only 26.9% of patients had examination under anesthesia for FIGO clinical staging.
CONCLUSION: There is a large discrepancy between the diagnostic tests recommended by FIGO and the actual tests used for cervical cancer staging, suggesting a need to reassess the relevance of the FIGO guidelines to current clinical practice in the United States.
INTRODUCTION
Invasive cervical cancer is the third most common gynecologic malignancy in the United States. It is estimated that in 2005, there will be 10,370 new cases of cervical cancer and 3,710 deaths from cervical cancer in the United States.[1] The initial choice of treatment largely depends on the cancer stage at the time of diagnosis, and consequently, accurate staging is essential for appropriate treatment. The International Federation of Gynecology and Obstetrics (FIGO) clinical staging system for cervical cancer has been universally adopted. While it calls for a number of conventional diagnostic tests, it does not include modern cross-sectional imaging. Comparison with surgical staging has shown that FIGO clinical staging results in understaging of up to 20% to 30% in stage IB, up to 23% in stage IIB, and nearly 40% in stage IIIB; it also produces overstaging of approximately 64% in stage IIIB.[2-6]
Previous patterns-of-care studies have shown a growing gap between the diagnostic tests recommended by FIGO and the diagnostic tests used in the United States. The FIGO guidelines recommend that staging be based on physical examination, colposcopy, lesion biopsy, chest radiography, cystoscopy, sigmoidoscopy, intravenous urogram (IVU), and barium enema (BE). But studies have reported that even 15 years ago, the use of conventional radiology examinations (ie, IVU, BE, and lymphangiography) was steadily decreasing, while the use of cross-sectional imaging was gaining acceptance in treatment planning.[7-10]
This prospective interdisciplinary clinical trial investigates the current utilization of diagnostic tests in the pretreatment work-up of invasive cervical cancer. It compares the results with those of prior patterns-of-care studies to identify changes that might have occurred or trends that might have continued over the past 15 years. [8-11]
PATIENTS AND METHODS
Patients
This is an interdisciplinary American College of Radiology Imaging Network (ACRIN)/Gynecologic Oncology Group (GOG) prospective clinical trial. Patients with documented cervical cancer clinical FIGO stage IB1 or greater who were scheduled for surgery were included in the study. All patients had FIGO clinical staging as the initial method of staging before enrollment. Each referring gynecologist determined which conventional FIGO clinical staging tests were to be used in what order and with what frequency. All studies performed during the patient's FIGO diagnostic work-up were recorded and subsequently tabulated. Computed tomography (CT) and magnetic resonance imaging (MRI) tests were performed on each patient after her entry onto the study, as the primary aim of the ACRIN/GOG study was to compare MRI and CT with each other and with clinical FIGO staging in the pretreatment evaluation of invasive cancer of the cervix, using surgico-pathologic findings as the reference standard. The results of the primary specific aim are reported separately (H.H., C.G., D.C., et al, unpublished observations, 2004).
Patient Inclusion Criteria
Only patients who had signed a study-specific informed consent form could be enrolled in the study. The patient inclusion criteria pertinent to the goals of this patterns-of-care analysis were: (1) biopsy-documented invasive cervical cancer; (2) clinical FIGO stage IB1 or greater assigned before enrollment; (3) FIGO clinical staging as the initial method of staging (additional diagnostic tests were performed, as needed, to complete the clinical pretreatment work-up); (4) clinical eligibility for surgery, with surgery intended at participating institutions. Allowable surgical procedures included hysterectomy (laparoscopic, transabdominal, or transvaginal), extrafascial transabdominal radical hysterectomy, or trachelectomy.
Exclusion Criteria
Patients unwilling to undergo contrast-enhanced CT and MRI were excluded from the study. Also excluded were pregnant patients and patients who, because of age, general medical or psychiatric condition, or physiologic status unrelated to the presence of cervical cancer could not give informed medical consent or were not considered surgical candidates. Also, patients with previous medical, surgical, or radiation treatment for invasive cervical cancer, and patients scheduled for a loop electrosurgical excision procedure (LEEP) procedure or only a cone biopsy were not eligible for this study.
Participating Institutions
Each participating institution was required to submit a protocol-specific application to ACRIN. Each imaging site had to be a GOG participant with a proven record of 20 surgical cases of gynecologic cancer per year. Imaging sites were required to have MRI (1.5T) and helical/spiral CT equipment (any vendor). In addition, each institution had to have at least two radiologists, a gynecologic oncologist, and a pathologist who would be designated to work on the study and would be committed to the study's goals.
Patient Accrual
Consecutive patients who met the selection criteria were recruited from the gynecologic oncology clinics of participating institutions. A patient's eligibility for the study was initially determined by the gynecologic oncologist. Eligibility was subsequently reviewed by the Data Management Center of ACRIN. All institutions had study-specific institutional review board approval.
Statistical Analysis
A total of 465 women were projected to enroll in this study over a period of 18 months. Assuming complete data on 440 cases, we expected that approximately 29% of cases would have a stage of IIB or higher. The sample size was selected to provide adequate power for the diagnostic performance end points of the study (reported elsewhere). The study began in March 2000. In consultation with the Data and Safety Monitoring Committee of ACRIN, the study was closed to further accrual in November 2002, with 208 patients enrolled. An important assumption in the design of the study was that CT and/or MRI results would not be incorporated in determining the clinical FIGO stage. However, practice patterns evolved during the period in which the study was designed and conducted, resulting in notably increased use of CT in the diagnostic work-up for cervical cancer staging. This change in practice patterns and slow patient accrual were the principal reasons for closing the study.
Data from the ACRIN/GOG protocol were collected, managed, and analyzed by the Biostatistics and Data Management Center of ACRIN. Comparative historical data on utilization were extracted from published reports as indicated in the results section.
Comparisons of rates of utilization of diagnostic procedures at each of the four time periods were performed using 2 tests for the hypothesis of equality of proportions.
RESULTS
Between March 1, 2000, and November 11, 2002, a total of 208 patients were enrolled by 25 participating institutions in the United States. One hundred ninety-nine of the enrolled patients met protocol eligibility criteria. The analysis in this report is based on data from the 197 patients who met the protocol eligibility criteria and for whom the necessary clinical assessment information was available.
In comparison with data previously reported by Montana et al,[8] our cohort included a larger proportion of younger women (34% below age 40 v 17% in 1989 and 89% below age 60 v 60% in 1989, 53% in 1983, and 60% in 1978; [Table 1]). The proportion of white women was similar to that reported in 1989 (68% v 67% in 1989), while the proportion of black women was smaller (15% v 21% in 1989; [Table 2]). The combined nonwhite proportion was similar to that of 1989 (32% v 33% in 1989) and substantially larger than the proportions reported in 1983 (21%) and 1978 (22%). The number of patients in the nonwhite/nonblack (other) category has continued to increase since 1978. Having risen from 3% to 4% to 12% in the first, second, and third surveys, respectively, the nonwhite/nonblack category comprised 17% of our total study population.[8]
FIGO-prescribed diagnostic procedures were performed at low rates in our study cohort ([Table 3]). Cystoscopy was performed on 8.1% of patients (16 of 197) and proctoscopy or sigmoidoscopy was performed on 8.6% of patients (17 of 197). The utilization of these diagnostic tests was significantly lower than in previous surveys ([Table 4]; P < .0001 in each instance) and has been declining since 1983. Note that in our study, a single category was used to report the performance of sigmoidoscopy and/or proctoscopy.
In the present study, intravenous urograms (IVUs) were recorded in only two of 197 patients (1.0%; [Table 4]). The use of IVUs has declined precipitously since 1983 (P < .0001), when it was reported at 91%.[8]
The use of barium enemas (BEs) has also continued to decline steeply ([Table 4]; P < .0001). No BEs were recorded in the present study, whereas the test was performed in 60% of patients in the 1983 study and in 32% of patients in the 1988 to 1989 study.[8]
No use of lymphangiography was recorded in the present study. This represents a notable change from the studies of 1978 (18%), 1983 (11%), and 1988 to 1989 (14%).[8]
DISCUSSION
Regarding demographic changes in the cohort, the rise in the proportion of patients in the nonwhite/nonblack (other) category likely reflects an overall increase in Hispanic patients with cervical cancer, as 28 of the 34 patients in the nonwhite/nonblack category were of Hispanic origin (14% of total). In turn, this increase probably reflects the overall rise in the Hispanic population of the United States. According to the US Census Bureau, in 2002, the Hispanic population became the largest minority in the United States, comprising about 13.5% of the total population.[12]
The prognosis of cervical carcinoma is primarily determined by the disease stage, primary tumor volume, and histologic grade.[7-10,13,14] The current FIGO staging system is based on clinical assessment, including findings from pelvic examination under anesthesia, colposcopy, sigmoidoscopy, lesion biopsy, endocervical curettage, hysteroscopy, cystoscopy, proctoscopy, IVU, BE, and x-rays of the lungs and skeleton.[15] Although evaluation of lymph node metastasis is an important prognostic factor and a determinant in treatment planning, it is not included in the clinical staging system.[15] Comparison with surgical staging has shown that FIGO clinical staging results in understaging of up to 20% to 30% in stage IB, up to 23% in stage IIB, and nearly 40% in stage IIIB; it also produces overstaging of approximately 64% in stage IIIB.[2-6]
Inaccuracy in clinical staging is predominantly because of difficulties in the evaluation of parametrial and pelvic sidewall invasion, bladder or rectal wall invasion, metastatic spread, and primary endocervical (endophytic) tumors.[3,6,13,14,16,17] Although numerous studies have shown that CT and MRI can aid in the evaluation of these prognostic factors,[11,18-21] modern cross-sectional imaging has not been incorporated into the clinical FIGO staging guidelines. Nevertheless, as knowledge of prognostic factors and of the value of cross-sectional imaging has spread, extended clinical staging has developed despite the official FIGO guidelines.[9,10] In a review of patterns-of-care studies conducted between 1978 and 1988 to 1989, Montana et al[8] reported decreases in the use of IVU (86% to 42%), BE (58% to 32%), and lymphangiography (18% to 14%), and a dramatic increase in the use of CT (6% to 70%). Similarly, a study by Russell et al[11] comparing patterns of cervical cancer care for 1984 and 1990 found that the use of cystoscopy, proctoscopy, IVU, BE, and lymphangiography had all declined by more than 25%, while the use of CT and MRI had risen sharply. Our study shows that the use of IVU, BE, and lymphangiography for cervical cancer staging has since declined even further (IVU to 1% and BE and lymphangiography to 0%)—to the point where it seems these tests no longer play a role in daily practice.
Nonsurgical evaluation of nodal metastasis is strongly advocated, as there is increased morbidity when surgical node dissection is combined with subsequent radiation therapy.[7,9,22,23] The use of lymphangiography in the pretreatment evaluation of cervical cancer used to be a source of controversy.[22] However, a meta-analysis study on the use of lymphangiography, CT, and MRI for the pretreatment evaluation of nodal disease in cervical cancer demonstrated positive predictive value (PPV) for lymphangiography ranging from as high as 80% to as low as 14% and better performance for both CT (PPV, 61%) and MRI (PPV, 66%).[23]
Although the current use of positron emission tomography (PET) in the initial evaluation of cervical cancer is still under investigation, it should be noted that PET can be used to assess nodal disease and tumor recurrence. In the detection of metastatic lymph nodes in patients with cervical cancer, PET has been reported to have a sensitivity of 91% and a specificity of 100%.[24] For the detection of recurrence, PET has been reported to have sensitivity of 85.7% to 90.3% and specificity of 76.1% to 86.7%.[25,26] None of the patients in our study underwent PET.
The primary aim of this ACRIN/GOG multicenter study was to compare CT and MRI with each other and with FIGO clinical staging in the pretreatment evaluation of invasive cervical cancer. The data analysis focused on the detection of cervical cancer stage IIB (ie, parametrial invasion) or above. CT and MRI performed similarly and had lower accuracy than in single-institution studies, with sensitivities of 42% and 53%, respectively, whereas FIGO clinical staging had unusually high accuracy, with a specificity of 99%. However, the temporal data showed that for 85% of the patients included in the data analysis, CT and/or MRI was performed before the submission of FIGO clinical staging forms. This temporal data and the limited usage of conventional FIGO-recommended tests provided indirect evidence that the exceptional performance of FIGO clinical staging was because of the inclusion of CT and/or MRI findings in the clinical staging process (H.H., C.G., D.C., et al, unpublished observations, 2004).
Thus, the primary analysis indirectly showed the value of cross-sectional imaging in assessing local stage and making the vital distinction between cervical cancer for which radical hysterectomy is appropriate (ie, stages < IIB) and cervical cancer for which it is not appropriate. The results came from a wide variety of academic and community imaging centers and should, therefore, reflect the actual value of imaging in daily practice. This suggests not only that cross-sectional imaging can already benefit patients with cervical cancer treated in a broad range of settings, but also that there is room for improvement of imaging results.
One limitation of our study was that because CT and MRI were required of all participants, it was not possible to compare our figures for CT and MRI usage with those from prior studies. The figures for pre-enrollment usage of CT and MRI in this study probably would not provide an accurate estimate of overall usage, since in clinical practice many of the patients who did not have pre-enrollment CT or MRI would likely have had CT or MRI some time before surgery even if they had not been enrolled in the clinical trial. Another limitation of our study was that clinicians might have decided against ordering some conventional studies, such as IVU, because of their awareness that the study subjects were to receive CT and MRI, which could have made the conventional tests appear redundant. Nevertheless, our results demonstrate that there is a large discrepancy between the diagnostic tests recommended by FIGO, including invasive tests such as cystoscopy and procto-sigmoidoscopy (obtained in only 8.1% and 8.6% of our study population, respectively), and the actual tests being used.
The lack of cross-sectional imaging recommendations in the FIGO cervical cancer staging guidelines probably results from the belief that staging methods should be universally available so that staging can serve as a standardized means of communication between institutions around the world. But is it best practice to maintain guidelines that are widely ignored, that are known to be inadequate, and that require a more youthfully distributed patient population to undergo numerous invasive procedures involving substantial radiation exposure? There is a body of literature showing the superiority of cross-sectional imaging—CT and MRI—over clinical staging.[18-20] FIGO staging is based on anatomy, and techniques continue to evolve for showing this anatomy more accurately. As CT and MRI techniques and training continue to develop, it is likely that accuracy for local staging will improve even further. We would like to propose that FIGO guidelines become more flexible and allow CT and MRI to be used when available. In developing countries, there is increasing availability of CT scanning. In the United States, IVUs are increasingly being replaced by CT scans and CT urography, making the idealistic goal of FIGO staging of comparing patients worldwide an elusive one.[27]
The thrust of medicine in the 21st century is toward minimally invasive diagnosis and treatment and optimal accuracy and effectiveness. In this context, the relevance of the current FIGO clinical staging guidelines in the United States may need to be reassessed.
Authors' Disclosures of Potential Conflicts of Interest
Although all authors completed the disclosure declaration, the following author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Acknowledgment
We thank the associates of the Biostatistics and Data Management Center of ACRIN for their work in data collection, management, and analysis. We also thank Ada Muellner for helping to write and edit the manuscript.
NOTES
Supported by Grants No. U01 CA07978 and U01 CA80098 from the National Cancer Institute.
ACRIN and GOG participated in the study design, data analysis, the critical revision of the report, and the decision to submit for publication.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Jemal A, Murray T, Ward E, et al: Cancer statistics, 2005. CA Cancer J Clin 55:10-30, 2005
Vidaurreta J, Bermudez A, di Paola G, et al: Laparoscopic staging in locally advanced cervical carcinoma: A new possible philosophy? Gynecol Oncol 75:366-371, 1999
Lagasse LD, Creasman WT, Shingleton HM, et al: Results and complications of operative staging in cervical cancer: Experience of the Gynecology Oncology Group. Gynecol Oncol 9:90-98, 1980
LaPolla JP, Schlaerth JB, Gaddis O, et al: The influence of surgical staging on the evaluation and treatment of patients with cervical carcinoma. Gynecol Oncol 24:194-199, 1986
Chan YM, Luesley DM: Screening, diagnosis, and staging of cervical cancer, in Gershenson DM, McGuire WP, Gore M, et al (eds): Gynecologic Cancer: Controversies in Management. Philadelphia, PA, Elsevier Churchill Livingstone, 2004, pp 31-45
Van Nagell JR Jr, Roddick JW Jr, Lowin DM: The staging of cervical cancer: Inevitable discrepancies between clinical staging and pathologic findings. Am J Obstet Gynecol 110:973-978, 1971
Hoskins WJ, Perez CA, Young RC: Gynecologic tumors, in DeVita VT, Hellman S, Rosenberg SA (eds): Cancer: Principles & Practice of Oncology (ed 3). Philadelphia, PA, Lippincott, 1989, pp 1114-1119
Montana GS, Hanlon AL, Brickner TJ, et al: Carcinoma of the cervix: Patterns of care studies: Review of 1978, 1983, and 1988-1989 surveys. Int J Radiat Oncol Biol Phys 32:1481-1486, 1995
Morrow CP, Curtin JP, Townsend DE: Tumors of the cervix, in Morrow CP, Curtin JP, Townsend DE (eds): Synopsis of Gynecologic Oncology. New York, NY, Churchill Livingstone, 1993, pp 111-152
Annual report on the results of treatment in gynecological cancer. Twenty-first volume. Statements of results obtained in patients treated in 1982 to 1986, inclusive 3 and 5-year survival up to 1990. Int J Gynaecol Obstet 36 Suppl:1-315, 1991
Russell AH, Shingleton HM, Jones WB, et al: Diagnostic assessments in patients with invasive cancer of the cervix: A national patterns of care study of the American College of Surgeons. Gynecol Oncol 63:159-165, 1996
US Census Bureau. American Factfinder. http://factfinder.census.gov/home/saff/main.html. Accessed May 13, 2004
Chung CK, Nahhas WA, Zaino R, et al: Histologic grade and lymph node metastasis in squamous cell carcinoma of the cervix. Gynecol Oncol 12:348-354, 1981
Zander J, Baltzer J, Lohe KJ, et al: Carcinoma of the cervix: An attempt to individualize treatment. Results of a 20-year cooperative study. Am J Obstet Gynecol 139:752-759, 1981
Pecorelli S, Odicino F: Cervical cancer staging. Cancer J 9:390-394, 2003
Delgado G, Bundy B, Zaino R, et al: Prospective surgical-pathological study of disease-free interval in patients with stage Ib squamous cell carcinoma of the cervix: A Gynecologic Oncology Group study. Gynecol Oncol 38:352-357, 1990
Dargent D, Frobert JL, Beau G: V factor (tumor volume) and T factor (FIGO classification) in the assessment of cervix cancer prognosis: The risk of lymph node spread. Gynecol Oncol 22:15-22, 1985
Hricak H, Lacey CG, Sandles LG, et al: Invasive cervical carcinoma: Comparison of MR imaging and surgical findings. Radiology 166:623-631, 1988
Kim SH, Choi BI, Han JK, et al: Preoperative staging of uterine cervical carcinoma: Comparison of CT and MR imaging in 99 patients. J Comput Assist Tomogr 17:633-640, 1993
Russell AH, Anderson M, Walter J, et al: The integration of computed tomography and magnetic resonance imaging in treatment planning for gynecologic cancer. Clin Obstet Gynecol 35:55-72, 1992
Subak LL, Hricak H, Powell CB, et al: Cervical carcinoma: Computed tomography and magnetic resonance imaging for preoperative staging. Obstet Gynecol 86:43-50, 1995
NIH Consensus Development Conference Statement: Cervical Cancer, Maryland: National Institutes of Health, April 1-3, 1996
Scheidler J, Hricak H, Yu KK, et al: Radiological evaluation of lymph node metastases in patients with cervical cancer. A meta-analysis. JAMA 278:1096-1101, 1997
Reinhardt MJ, Ehritt-Braun C, Vogelgesang D, et al: Metastatic lymph nodes in patients with cervical cancer: Detection with MR imaging and FDG PET. Radiology 218:776-782, 2001
Ryu SY, Kim MH, Choi SC, et al: Detection of early recurrence with 18F-FDG PET in patients with cervical cancer. J Nucl Med 44:347-352, 2003
Havrilesky LJ, Wong TZ, Secord AA, et al: The role of PET scanning in the detection of recurrent cervical cancer. Gynecol Oncol 90:186-190, 2003
Dunnick NR, Sandler CM, Newhouse JH, et al (eds): Textbook of Uroradiology. Philadelphia, PA, Lippincott Williams & Wilkins, 2001, pp 49-72
Submitted November 12, 2004; accepted July 6, 2005.(Marco A. Amendola, Hedvig)
Department of Radiology, Memorial Sloan-Kettering Cancer Center, New York, NY
Department of Radiology, Thomas Jefferson University Hospital, Philadelphia, PA
Center for Statistical Sciences, Brown University, Providence, RI
Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY
Department of Medical Oncology, Mayo Clinic, Rochester, MN
College of Medicine Division of Gynecologic Oncology, Lee Moffitt Cancer Center, Tampa, FL
ABSTRACT
PURPOSE: To review the current utilization of diagnostic tests prescribed by the International Federation of Gynecology and Obstetrics (FIGO) clinical staging guidelines in the pretreatment work-up of invasive cervical cancer, and to compare the data with those of previous patterns of care studies.
PATIENTS AND METHODS: This interdisciplinary American College of Radiology Imaging Network/Gynecologic Oncology Group prospective clinical trial was conducted between March 1, 2000, and November 11, 2002. Twenty-five participating institutions, all from the United States, enrolled a total of 208 patients. Only patients scheduled for surgery with biopsy-confirmed cervical cancer of clinical FIGO stage IB or higher were eligible. The patterns of care data analysis was based on 197 patients who met all inclusion criteria. The conventional FIGO-recommended tests used for pre-enrollment FIGO clinical stage classification were at the discretion of the treating physician; overall frequency of use was tabulated for each test.
RESULTS: Use of cystoscopy (8.1%) and sigmoidoscopy or proctoscopy (8.6%) was significantly lower than in 1988 to 1989 (P < .0001 in each instance). Intravenous urography was used in only 1% of patients as compared with 42% in 1988 to 1989 and 91% in 1983. No patient included in the data analysis had barium enema or lymphangiography. Only 26.9% of patients had examination under anesthesia for FIGO clinical staging.
CONCLUSION: There is a large discrepancy between the diagnostic tests recommended by FIGO and the actual tests used for cervical cancer staging, suggesting a need to reassess the relevance of the FIGO guidelines to current clinical practice in the United States.
INTRODUCTION
Invasive cervical cancer is the third most common gynecologic malignancy in the United States. It is estimated that in 2005, there will be 10,370 new cases of cervical cancer and 3,710 deaths from cervical cancer in the United States.[1] The initial choice of treatment largely depends on the cancer stage at the time of diagnosis, and consequently, accurate staging is essential for appropriate treatment. The International Federation of Gynecology and Obstetrics (FIGO) clinical staging system for cervical cancer has been universally adopted. While it calls for a number of conventional diagnostic tests, it does not include modern cross-sectional imaging. Comparison with surgical staging has shown that FIGO clinical staging results in understaging of up to 20% to 30% in stage IB, up to 23% in stage IIB, and nearly 40% in stage IIIB; it also produces overstaging of approximately 64% in stage IIIB.[2-6]
Previous patterns-of-care studies have shown a growing gap between the diagnostic tests recommended by FIGO and the diagnostic tests used in the United States. The FIGO guidelines recommend that staging be based on physical examination, colposcopy, lesion biopsy, chest radiography, cystoscopy, sigmoidoscopy, intravenous urogram (IVU), and barium enema (BE). But studies have reported that even 15 years ago, the use of conventional radiology examinations (ie, IVU, BE, and lymphangiography) was steadily decreasing, while the use of cross-sectional imaging was gaining acceptance in treatment planning.[7-10]
This prospective interdisciplinary clinical trial investigates the current utilization of diagnostic tests in the pretreatment work-up of invasive cervical cancer. It compares the results with those of prior patterns-of-care studies to identify changes that might have occurred or trends that might have continued over the past 15 years. [8-11]
PATIENTS AND METHODS
Patients
This is an interdisciplinary American College of Radiology Imaging Network (ACRIN)/Gynecologic Oncology Group (GOG) prospective clinical trial. Patients with documented cervical cancer clinical FIGO stage IB1 or greater who were scheduled for surgery were included in the study. All patients had FIGO clinical staging as the initial method of staging before enrollment. Each referring gynecologist determined which conventional FIGO clinical staging tests were to be used in what order and with what frequency. All studies performed during the patient's FIGO diagnostic work-up were recorded and subsequently tabulated. Computed tomography (CT) and magnetic resonance imaging (MRI) tests were performed on each patient after her entry onto the study, as the primary aim of the ACRIN/GOG study was to compare MRI and CT with each other and with clinical FIGO staging in the pretreatment evaluation of invasive cancer of the cervix, using surgico-pathologic findings as the reference standard. The results of the primary specific aim are reported separately (H.H., C.G., D.C., et al, unpublished observations, 2004).
Patient Inclusion Criteria
Only patients who had signed a study-specific informed consent form could be enrolled in the study. The patient inclusion criteria pertinent to the goals of this patterns-of-care analysis were: (1) biopsy-documented invasive cervical cancer; (2) clinical FIGO stage IB1 or greater assigned before enrollment; (3) FIGO clinical staging as the initial method of staging (additional diagnostic tests were performed, as needed, to complete the clinical pretreatment work-up); (4) clinical eligibility for surgery, with surgery intended at participating institutions. Allowable surgical procedures included hysterectomy (laparoscopic, transabdominal, or transvaginal), extrafascial transabdominal radical hysterectomy, or trachelectomy.
Exclusion Criteria
Patients unwilling to undergo contrast-enhanced CT and MRI were excluded from the study. Also excluded were pregnant patients and patients who, because of age, general medical or psychiatric condition, or physiologic status unrelated to the presence of cervical cancer could not give informed medical consent or were not considered surgical candidates. Also, patients with previous medical, surgical, or radiation treatment for invasive cervical cancer, and patients scheduled for a loop electrosurgical excision procedure (LEEP) procedure or only a cone biopsy were not eligible for this study.
Participating Institutions
Each participating institution was required to submit a protocol-specific application to ACRIN. Each imaging site had to be a GOG participant with a proven record of 20 surgical cases of gynecologic cancer per year. Imaging sites were required to have MRI (1.5T) and helical/spiral CT equipment (any vendor). In addition, each institution had to have at least two radiologists, a gynecologic oncologist, and a pathologist who would be designated to work on the study and would be committed to the study's goals.
Patient Accrual
Consecutive patients who met the selection criteria were recruited from the gynecologic oncology clinics of participating institutions. A patient's eligibility for the study was initially determined by the gynecologic oncologist. Eligibility was subsequently reviewed by the Data Management Center of ACRIN. All institutions had study-specific institutional review board approval.
Statistical Analysis
A total of 465 women were projected to enroll in this study over a period of 18 months. Assuming complete data on 440 cases, we expected that approximately 29% of cases would have a stage of IIB or higher. The sample size was selected to provide adequate power for the diagnostic performance end points of the study (reported elsewhere). The study began in March 2000. In consultation with the Data and Safety Monitoring Committee of ACRIN, the study was closed to further accrual in November 2002, with 208 patients enrolled. An important assumption in the design of the study was that CT and/or MRI results would not be incorporated in determining the clinical FIGO stage. However, practice patterns evolved during the period in which the study was designed and conducted, resulting in notably increased use of CT in the diagnostic work-up for cervical cancer staging. This change in practice patterns and slow patient accrual were the principal reasons for closing the study.
Data from the ACRIN/GOG protocol were collected, managed, and analyzed by the Biostatistics and Data Management Center of ACRIN. Comparative historical data on utilization were extracted from published reports as indicated in the results section.
Comparisons of rates of utilization of diagnostic procedures at each of the four time periods were performed using 2 tests for the hypothesis of equality of proportions.
RESULTS
Between March 1, 2000, and November 11, 2002, a total of 208 patients were enrolled by 25 participating institutions in the United States. One hundred ninety-nine of the enrolled patients met protocol eligibility criteria. The analysis in this report is based on data from the 197 patients who met the protocol eligibility criteria and for whom the necessary clinical assessment information was available.
In comparison with data previously reported by Montana et al,[8] our cohort included a larger proportion of younger women (34% below age 40 v 17% in 1989 and 89% below age 60 v 60% in 1989, 53% in 1983, and 60% in 1978; [Table 1]). The proportion of white women was similar to that reported in 1989 (68% v 67% in 1989), while the proportion of black women was smaller (15% v 21% in 1989; [Table 2]). The combined nonwhite proportion was similar to that of 1989 (32% v 33% in 1989) and substantially larger than the proportions reported in 1983 (21%) and 1978 (22%). The number of patients in the nonwhite/nonblack (other) category has continued to increase since 1978. Having risen from 3% to 4% to 12% in the first, second, and third surveys, respectively, the nonwhite/nonblack category comprised 17% of our total study population.[8]
FIGO-prescribed diagnostic procedures were performed at low rates in our study cohort ([Table 3]). Cystoscopy was performed on 8.1% of patients (16 of 197) and proctoscopy or sigmoidoscopy was performed on 8.6% of patients (17 of 197). The utilization of these diagnostic tests was significantly lower than in previous surveys ([Table 4]; P < .0001 in each instance) and has been declining since 1983. Note that in our study, a single category was used to report the performance of sigmoidoscopy and/or proctoscopy.
In the present study, intravenous urograms (IVUs) were recorded in only two of 197 patients (1.0%; [Table 4]). The use of IVUs has declined precipitously since 1983 (P < .0001), when it was reported at 91%.[8]
The use of barium enemas (BEs) has also continued to decline steeply ([Table 4]; P < .0001). No BEs were recorded in the present study, whereas the test was performed in 60% of patients in the 1983 study and in 32% of patients in the 1988 to 1989 study.[8]
No use of lymphangiography was recorded in the present study. This represents a notable change from the studies of 1978 (18%), 1983 (11%), and 1988 to 1989 (14%).[8]
DISCUSSION
Regarding demographic changes in the cohort, the rise in the proportion of patients in the nonwhite/nonblack (other) category likely reflects an overall increase in Hispanic patients with cervical cancer, as 28 of the 34 patients in the nonwhite/nonblack category were of Hispanic origin (14% of total). In turn, this increase probably reflects the overall rise in the Hispanic population of the United States. According to the US Census Bureau, in 2002, the Hispanic population became the largest minority in the United States, comprising about 13.5% of the total population.[12]
The prognosis of cervical carcinoma is primarily determined by the disease stage, primary tumor volume, and histologic grade.[7-10,13,14] The current FIGO staging system is based on clinical assessment, including findings from pelvic examination under anesthesia, colposcopy, sigmoidoscopy, lesion biopsy, endocervical curettage, hysteroscopy, cystoscopy, proctoscopy, IVU, BE, and x-rays of the lungs and skeleton.[15] Although evaluation of lymph node metastasis is an important prognostic factor and a determinant in treatment planning, it is not included in the clinical staging system.[15] Comparison with surgical staging has shown that FIGO clinical staging results in understaging of up to 20% to 30% in stage IB, up to 23% in stage IIB, and nearly 40% in stage IIIB; it also produces overstaging of approximately 64% in stage IIIB.[2-6]
Inaccuracy in clinical staging is predominantly because of difficulties in the evaluation of parametrial and pelvic sidewall invasion, bladder or rectal wall invasion, metastatic spread, and primary endocervical (endophytic) tumors.[3,6,13,14,16,17] Although numerous studies have shown that CT and MRI can aid in the evaluation of these prognostic factors,[11,18-21] modern cross-sectional imaging has not been incorporated into the clinical FIGO staging guidelines. Nevertheless, as knowledge of prognostic factors and of the value of cross-sectional imaging has spread, extended clinical staging has developed despite the official FIGO guidelines.[9,10] In a review of patterns-of-care studies conducted between 1978 and 1988 to 1989, Montana et al[8] reported decreases in the use of IVU (86% to 42%), BE (58% to 32%), and lymphangiography (18% to 14%), and a dramatic increase in the use of CT (6% to 70%). Similarly, a study by Russell et al[11] comparing patterns of cervical cancer care for 1984 and 1990 found that the use of cystoscopy, proctoscopy, IVU, BE, and lymphangiography had all declined by more than 25%, while the use of CT and MRI had risen sharply. Our study shows that the use of IVU, BE, and lymphangiography for cervical cancer staging has since declined even further (IVU to 1% and BE and lymphangiography to 0%)—to the point where it seems these tests no longer play a role in daily practice.
Nonsurgical evaluation of nodal metastasis is strongly advocated, as there is increased morbidity when surgical node dissection is combined with subsequent radiation therapy.[7,9,22,23] The use of lymphangiography in the pretreatment evaluation of cervical cancer used to be a source of controversy.[22] However, a meta-analysis study on the use of lymphangiography, CT, and MRI for the pretreatment evaluation of nodal disease in cervical cancer demonstrated positive predictive value (PPV) for lymphangiography ranging from as high as 80% to as low as 14% and better performance for both CT (PPV, 61%) and MRI (PPV, 66%).[23]
Although the current use of positron emission tomography (PET) in the initial evaluation of cervical cancer is still under investigation, it should be noted that PET can be used to assess nodal disease and tumor recurrence. In the detection of metastatic lymph nodes in patients with cervical cancer, PET has been reported to have a sensitivity of 91% and a specificity of 100%.[24] For the detection of recurrence, PET has been reported to have sensitivity of 85.7% to 90.3% and specificity of 76.1% to 86.7%.[25,26] None of the patients in our study underwent PET.
The primary aim of this ACRIN/GOG multicenter study was to compare CT and MRI with each other and with FIGO clinical staging in the pretreatment evaluation of invasive cervical cancer. The data analysis focused on the detection of cervical cancer stage IIB (ie, parametrial invasion) or above. CT and MRI performed similarly and had lower accuracy than in single-institution studies, with sensitivities of 42% and 53%, respectively, whereas FIGO clinical staging had unusually high accuracy, with a specificity of 99%. However, the temporal data showed that for 85% of the patients included in the data analysis, CT and/or MRI was performed before the submission of FIGO clinical staging forms. This temporal data and the limited usage of conventional FIGO-recommended tests provided indirect evidence that the exceptional performance of FIGO clinical staging was because of the inclusion of CT and/or MRI findings in the clinical staging process (H.H., C.G., D.C., et al, unpublished observations, 2004).
Thus, the primary analysis indirectly showed the value of cross-sectional imaging in assessing local stage and making the vital distinction between cervical cancer for which radical hysterectomy is appropriate (ie, stages < IIB) and cervical cancer for which it is not appropriate. The results came from a wide variety of academic and community imaging centers and should, therefore, reflect the actual value of imaging in daily practice. This suggests not only that cross-sectional imaging can already benefit patients with cervical cancer treated in a broad range of settings, but also that there is room for improvement of imaging results.
One limitation of our study was that because CT and MRI were required of all participants, it was not possible to compare our figures for CT and MRI usage with those from prior studies. The figures for pre-enrollment usage of CT and MRI in this study probably would not provide an accurate estimate of overall usage, since in clinical practice many of the patients who did not have pre-enrollment CT or MRI would likely have had CT or MRI some time before surgery even if they had not been enrolled in the clinical trial. Another limitation of our study was that clinicians might have decided against ordering some conventional studies, such as IVU, because of their awareness that the study subjects were to receive CT and MRI, which could have made the conventional tests appear redundant. Nevertheless, our results demonstrate that there is a large discrepancy between the diagnostic tests recommended by FIGO, including invasive tests such as cystoscopy and procto-sigmoidoscopy (obtained in only 8.1% and 8.6% of our study population, respectively), and the actual tests being used.
The lack of cross-sectional imaging recommendations in the FIGO cervical cancer staging guidelines probably results from the belief that staging methods should be universally available so that staging can serve as a standardized means of communication between institutions around the world. But is it best practice to maintain guidelines that are widely ignored, that are known to be inadequate, and that require a more youthfully distributed patient population to undergo numerous invasive procedures involving substantial radiation exposure? There is a body of literature showing the superiority of cross-sectional imaging—CT and MRI—over clinical staging.[18-20] FIGO staging is based on anatomy, and techniques continue to evolve for showing this anatomy more accurately. As CT and MRI techniques and training continue to develop, it is likely that accuracy for local staging will improve even further. We would like to propose that FIGO guidelines become more flexible and allow CT and MRI to be used when available. In developing countries, there is increasing availability of CT scanning. In the United States, IVUs are increasingly being replaced by CT scans and CT urography, making the idealistic goal of FIGO staging of comparing patients worldwide an elusive one.[27]
The thrust of medicine in the 21st century is toward minimally invasive diagnosis and treatment and optimal accuracy and effectiveness. In this context, the relevance of the current FIGO clinical staging guidelines in the United States may need to be reassessed.
Authors' Disclosures of Potential Conflicts of Interest
Although all authors completed the disclosure declaration, the following author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Acknowledgment
We thank the associates of the Biostatistics and Data Management Center of ACRIN for their work in data collection, management, and analysis. We also thank Ada Muellner for helping to write and edit the manuscript.
NOTES
Supported by Grants No. U01 CA07978 and U01 CA80098 from the National Cancer Institute.
ACRIN and GOG participated in the study design, data analysis, the critical revision of the report, and the decision to submit for publication.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Jemal A, Murray T, Ward E, et al: Cancer statistics, 2005. CA Cancer J Clin 55:10-30, 2005
Vidaurreta J, Bermudez A, di Paola G, et al: Laparoscopic staging in locally advanced cervical carcinoma: A new possible philosophy? Gynecol Oncol 75:366-371, 1999
Lagasse LD, Creasman WT, Shingleton HM, et al: Results and complications of operative staging in cervical cancer: Experience of the Gynecology Oncology Group. Gynecol Oncol 9:90-98, 1980
LaPolla JP, Schlaerth JB, Gaddis O, et al: The influence of surgical staging on the evaluation and treatment of patients with cervical carcinoma. Gynecol Oncol 24:194-199, 1986
Chan YM, Luesley DM: Screening, diagnosis, and staging of cervical cancer, in Gershenson DM, McGuire WP, Gore M, et al (eds): Gynecologic Cancer: Controversies in Management. Philadelphia, PA, Elsevier Churchill Livingstone, 2004, pp 31-45
Van Nagell JR Jr, Roddick JW Jr, Lowin DM: The staging of cervical cancer: Inevitable discrepancies between clinical staging and pathologic findings. Am J Obstet Gynecol 110:973-978, 1971
Hoskins WJ, Perez CA, Young RC: Gynecologic tumors, in DeVita VT, Hellman S, Rosenberg SA (eds): Cancer: Principles & Practice of Oncology (ed 3). Philadelphia, PA, Lippincott, 1989, pp 1114-1119
Montana GS, Hanlon AL, Brickner TJ, et al: Carcinoma of the cervix: Patterns of care studies: Review of 1978, 1983, and 1988-1989 surveys. Int J Radiat Oncol Biol Phys 32:1481-1486, 1995
Morrow CP, Curtin JP, Townsend DE: Tumors of the cervix, in Morrow CP, Curtin JP, Townsend DE (eds): Synopsis of Gynecologic Oncology. New York, NY, Churchill Livingstone, 1993, pp 111-152
Annual report on the results of treatment in gynecological cancer. Twenty-first volume. Statements of results obtained in patients treated in 1982 to 1986, inclusive 3 and 5-year survival up to 1990. Int J Gynaecol Obstet 36 Suppl:1-315, 1991
Russell AH, Shingleton HM, Jones WB, et al: Diagnostic assessments in patients with invasive cancer of the cervix: A national patterns of care study of the American College of Surgeons. Gynecol Oncol 63:159-165, 1996
US Census Bureau. American Factfinder. http://factfinder.census.gov/home/saff/main.html. Accessed May 13, 2004
Chung CK, Nahhas WA, Zaino R, et al: Histologic grade and lymph node metastasis in squamous cell carcinoma of the cervix. Gynecol Oncol 12:348-354, 1981
Zander J, Baltzer J, Lohe KJ, et al: Carcinoma of the cervix: An attempt to individualize treatment. Results of a 20-year cooperative study. Am J Obstet Gynecol 139:752-759, 1981
Pecorelli S, Odicino F: Cervical cancer staging. Cancer J 9:390-394, 2003
Delgado G, Bundy B, Zaino R, et al: Prospective surgical-pathological study of disease-free interval in patients with stage Ib squamous cell carcinoma of the cervix: A Gynecologic Oncology Group study. Gynecol Oncol 38:352-357, 1990
Dargent D, Frobert JL, Beau G: V factor (tumor volume) and T factor (FIGO classification) in the assessment of cervix cancer prognosis: The risk of lymph node spread. Gynecol Oncol 22:15-22, 1985
Hricak H, Lacey CG, Sandles LG, et al: Invasive cervical carcinoma: Comparison of MR imaging and surgical findings. Radiology 166:623-631, 1988
Kim SH, Choi BI, Han JK, et al: Preoperative staging of uterine cervical carcinoma: Comparison of CT and MR imaging in 99 patients. J Comput Assist Tomogr 17:633-640, 1993
Russell AH, Anderson M, Walter J, et al: The integration of computed tomography and magnetic resonance imaging in treatment planning for gynecologic cancer. Clin Obstet Gynecol 35:55-72, 1992
Subak LL, Hricak H, Powell CB, et al: Cervical carcinoma: Computed tomography and magnetic resonance imaging for preoperative staging. Obstet Gynecol 86:43-50, 1995
NIH Consensus Development Conference Statement: Cervical Cancer, Maryland: National Institutes of Health, April 1-3, 1996
Scheidler J, Hricak H, Yu KK, et al: Radiological evaluation of lymph node metastases in patients with cervical cancer. A meta-analysis. JAMA 278:1096-1101, 1997
Reinhardt MJ, Ehritt-Braun C, Vogelgesang D, et al: Metastatic lymph nodes in patients with cervical cancer: Detection with MR imaging and FDG PET. Radiology 218:776-782, 2001
Ryu SY, Kim MH, Choi SC, et al: Detection of early recurrence with 18F-FDG PET in patients with cervical cancer. J Nucl Med 44:347-352, 2003
Havrilesky LJ, Wong TZ, Secord AA, et al: The role of PET scanning in the detection of recurrent cervical cancer. Gynecol Oncol 90:186-190, 2003
Dunnick NR, Sandler CM, Newhouse JH, et al (eds): Textbook of Uroradiology. Philadelphia, PA, Lippincott Williams & Wilkins, 2001, pp 49-72
Submitted November 12, 2004; accepted July 6, 2005.(Marco A. Amendola, Hedvig)