Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Postema, S.
Right arrow Articles by Trimbos, J. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Postema, S.
Right arrow Articles by Trimbos, J. B.
(Radiology. 1999;210:217-220.)
© RSNA, 1999

Genitourinary Imaging

Cervical Carcinoma: Can Dynamic Contrast-enhanced MR Imaging Help Predict Tumor Aggressiveness?

Sandra Postema, MD1, Peter M. T. Pattynama, MD2,1, Carla S. P. van Rijswijk, MD1 and J. Baptist Trimbos, MD2

1 Departments of Radiology (S.P., P.M.T.P., C.S.P.v.R.)
2 Gynecology (J.B.T.), Leiden University Medical Center, the Netherlands.


   Abstract
TOP
 Abstract
Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
PURPOSE: To determine whether fast dynamic contrast agent–enhanced magnetic resonance (MR) imaging can demonstrate tumor aggressiveness of cervical carcinoma in patients who are eligible for surgical treatment.

MATERIALS AND METHODS: Dynamic contrast-enhanced MR imaging of cervical carcinoma was performed in 82 consecutive patients with stage I or IIA disease who were referred for radical hysterectomy. The maximum slope and amplitude of dynamic first-pass contrast enhancement were quantified. These parameters were correlated with histologic measures of tumor aggressiveness (tumor invasion depth, pelvic lymph node status).

RESULTS: The analysis was based on tumors in 62 patients: 30 aggressive and 32 relatively nonaggressive tumors. Twenty patients were excluded from analysis owing to insufficient surgical data, tumor too small for accurate assessment, or technical problems. There were no significant differences between aggressive and nonaggressive tumors in terms of the first-pass contrast-enhancement parameters of slope (2.0 vs 2.1 arbitrary signal intensity units per second, P > .5) or amplitude (24.8 vs 27.8 arbitrary units, P > .2).

CONCLUSION: Dynamic contrast-enhanced MR imaging does not facilitate differentiation between aggressive and nonaggressive tumors and therefore has no clinical role in assisting in treatment decisions in patients who are candidates for radical hysterectomy.

Index terms: Magnetic resonance (MR), treatment planning, 854.32, 854.33 • Uterine neoplasms, 854.32 • Uterine neoplasms, metastases, 992.33 • Uterus, MR, 854.121412, 854.12143


   Introduction
TOP
 Abstract
 Introduction
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Magnetic resonance (MR) imaging has been shown to be helpful in tumor staging of uterine cervical carcinoma. An assessment of tumor ingrowth into the surrounding tissues helps guide the choice of treatment (1–4). For example, at many centers, carcinoma confined to the cervix will be treated with surgery, whereas radiation therapy will be administered whenever the tumor has invaded the parametria, the rectum, or the bladder wall. Additional postoperative radiation therapy will be administered when tumor invasion depth proves to be 15 mm or more and/or positive pelvic lymph nodes are found at surgery, because these findings are considered to be indicative of a high risk for tumor recurrence (5,6). Of importance, if one could distinguish before surgery those patients with deep tumor invasion or lymph node metastases, primary radiation therapy could be considered so that such patients would not have to undergo a double treatment with all the related morbidity.

It has recently been reported that MR imaging contrast agents and dynamic MR imaging may play roles in this respect. Hawighorst et al (7) observed that the tumors associated with a more pronounced degree of tumor angiogenesis have a markedly shorter exchange rate constant; that is, such tumors show more rapid first-pass contrast agent enhancement at dynamic MR imaging. Because tumor angiogenesis has been implicated as a predictor for tumor recurrence (8), fast and/or intense first-pass contrast enhancement of a tumor would correspond to more aggressive behavior (7,9). In theory, dynamic MR imaging could be used for the preoperative identification of aggressive tumor types, which could then be treated with primary radiation therapy. Dynamic MR imaging could thus help prevent unnecessary surgery in patients who, on the basis of conventional International Federation of Gynecology and Obstetrics (FIGO) staging criteria, would undergo surgery. The purpose of this study was to validate this hypothesis.

We undertook a study to determine whether such use of dynamic MR imaging would be feasible in clinical practice. We assessed whether, in the group of patients who were eligible for primary surgical treatment, contrast-enhanced dynamic MR imaging could be used for reliable determination of which patients had more aggressive cervical carcinomas, that is, tumors that, according to generally accepted criteria, should be treated with radiation therapy.


   MATERIALS AND METHODS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Study Design
Our study population consisted of a group of consecutive patients who, according to conventional FIGO criteria, would be treated with primary surgery (stage IA, IB, or IIA disease). Dynamic MR imaging was performed within 2 weeks before surgery. Results from postoperative histologic analyses of the hysterectomy specimen and lymph node biopsy specimens were used to determine the extent of tumor invasion (>=15 mm or <15 mm), the presence of pelvic metastatic lymph node involvement, or both. These two criteria are generally accepted as indicative of the risk for tumor recurrence (5,6). The analysts of the histopathologic material were blinded to the MR imaging data.

We then analyzed whether dynamic contrast-enhanced MR imaging could help distinguish, before surgery, patients at high risk on the basis of the quantification of first-pass contrast enhancement. The MR imaging results did not play a role in the decision about the type of treatment. The study was approved by the ethical review board of our institution, and informed consent was obtained from all patients.

Patients
The study group comprised 82 consecutive patients aged 28–85 years (mean age, 47.9 years) who were referred to the gynecology department at our hospital from February 1995 to July 1997 with histologically proved invasive cervical carcinoma, and who, on the basis of clinical FIGO staging, were to undergo surgical treatment (clinical stages IA, IB, IIA).

MR Imaging Protocol
MR imaging was performed with a 1.5-T system (Gyroscan NT-15; Philips Medical Systems, Best, the Netherlands) equipped with a body coil. For dynamic MR imaging, two contiguous, transverse, 7-mm-thick sections were obtained through the cervical tumor, with a temporal resolution of one image every two seconds. A T1-weighted fast gradient-recalled-echo sequence was used (15/7 [repetition time msec/echo time msec]; flip angle, 40°; one signal acquired; matrix size, 256 x 128; field of view, 45 cm). A bolus of 0.1 mmol of gadopentetate dimeglumine (Magnevist; Berlex-Schering, Berlin, Germany) per kilogram of body weight was manually injected; a flush with saline solution was delivered after the fourth gradient-recalled-echo sequence. The contrast-enhanced dynamic images were subtracted from the third unenhanced dynamic image (in which the spins had reached equilibrium saturation after having been exposed to the repetitive {alpha} pulses in the gradient-recalled-echo sequence). The imaging protocol included acquisition of T2-weighted fast spin-echo images in the transverse and sagittal planes (2,500/120 [effective echo time]; echo train length, 18; matrix, 256 x 256; field of view, 22 cm; no fat suppression; eight signals acquired).

Quantification of First-Pass Contrast Enhancement
By using the morphologic information provided by the T2-weighted images, regions of interest were drawn on the unenhanced dynamic MR images inside the cervical carcinoma and skeletal muscle (gluteal muscles). The time–signal intensity curves in the regions of interest were computed and were displayed graphically. All curves had a fast, approximately linear rise followed by a second phase of a plateau without substantial washout (Figure). We used automated computer software to record the following parameters at the quantification of the first-pass curves: (a) maximum slope of contrast enhancement (a linear fit on the signal intensity curve derived from four subsequent images was used to reduce the influence of random noise; the linear fit was based on the least-squares analysis method) and (b) amplitude of contrast enhancement (the maximum signal intensity reached in the subtracted image). The maximum slope and the amplitude were expressed in arbitrary signal intensity units per second and arbitrary signal intensity units, respectively. To eliminate variations due to differences in contrast agent injection velocity, heart rate, and cardiac output, we constructed an additional data set in which the values of the maximum slope and the amplitude were scaled to the corresponding values derived from normal skeletal muscle. This was accomplished by dividing the slope and amplitude values of the cervical carcinoma by the corresponding values of normal skeletal muscle in the same patient. All measurements were made by an observer (C.S.P.v.R.) who was unaware of the surgical-histopathologic findings.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 1. Time–signal intensity curves derived from fast dynamic contrast-enhanced MR imaging of cervical carcinoma and skeletal muscle in one patient. The curve for cervical carcinoma shows the steep positive slope and high amplitude of contrast enhancement as compared with the slope and amplitude of the curve for normal skeletal muscle. max. = maximum.

 
Statistical Analyses
On the basis of the histologic analysis of tumor invasion depth (<15 mm vs >=15 mm) and the presence of lymph node metastases, we formed two subgroups: relatively nonaggressive tumors (tumor invasion depth < 15 mm, no lymph node metastases) and aggressive tumors (tumor invasion depth >= 15 mm, presence of lymph node metastases, or both). For each tumor, we recorded maximum slope and amplitude for the absolute tumor signal intensity values and for the relative tumor values (ie, scaled to normal skeletal muscle). We used two-tailed unpaired Student t tests to test for significant differences between the two subgroups, with a P value of .05 or less indicative of such differences.


   RESULTS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
The disease stage, pelvic nodal status, and tumor invasion depth in the 82 patients who, on the basis of clinical staging criteria, were candidates for surgery are summarized in Tables 1 and 2. Twelve patients had a stage IA tumor, 48 had a stage IB tumor, seven had a stage IIA tumor, six had a stage IIB tumor, and four had a stage IVA tumor. In five patients, the exact tumor stage could not be determined, because no hysterectomy was performed or because of surgical technical reasons in the presence of proved pelvic lymph node metastases. The disease stages determined after surgery may differ from the preoperative clinical FIGO stages. For our analysis, we used the preoperative clinical FIGO stages.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Tumor Stage, Nodal Status, and Tumor Invasion Depth in 62 Patients Included in the Analysis
 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Tumor Stage, Nodal Status, and Tumor Invasion Depth in 20 Patients Excluded from the Analysis
 
In 77 patients, we could determine tumor aggressiveness on the basis of nodal status and histologic depth of tumor invasion. In the remaining five patients, it was not possible to determine tumor aggressiveness because insufficient data were available. Of the 10 patients who, in the end, had a more advanced stage of disease, six had parametrial involvement, and four had bladder wall infiltration.

Twenty patients were excluded from the analyses (Table 2). Although MR imaging was performed in all 82 patients, dynamic contrast-enhanced MR imaging was not completed in three due to technical problems. In the 12 patients with stage IA disease, no tumor could be visualized on the T2-weighted MR images. The five patients in whom tumor aggressiveness could not be assessed because of insufficient data were also excluded from the analysis. Overall, a correlation between dynamic MR imaging results and tumor aggressiveness could be established in 62 patients, including 32 with nonaggressive tumors and 30 with aggressive tumors (Table 1).

The first-pass dynamic contrast-enhancement parameters in these 62 patients are summarized in Table 3. Faster and more intense first-pass contrast-enhancement was seen in tumor than in skeletal muscle in all 62 patients (Figure). There was a significant difference between the absolute slope and amplitude mean values for tumor and skeletal muscle (slope, 2.1 arbitrary units per second ± 0.8 vs 0.4 units per second ± 0.2, P << .001; amplitude, 26.4 units ± 9.3 vs 4.2 units ± 2.2, P << .001). We found no differences between aggressive and nonaggressive tumors in terms of absolute maximum slope (2.0 units per second ± 0.7 vs 2.1 units per second ± 0.9, P > .5) or amplitude (24.8 units ± 7.7 vs 27.8 units ± 10.5, P > .2) or in terms of relative values scaled to skeletal muscle (slope, 6.0 units per second ± 3.6 vs 5.5 units per second ± 3.6, P > .5; amplitude, 9.8 units ± 9.3 vs 7.6 units ± 6.9, P > .5). Dynamic MR images, therefore, reliably showed the difference between tumor tissue and muscle but did not allow differentiation between aggressive and nonaggressive tumors in our patient group.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Dynamic MR Imaging Signal Intensity Parameters: Absolute Values and Relative Values Scaled to Normal Skeletal Muscle
 

   DISCUSSION
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
In recent studies (7,10), it has been shown that contrast-enhanced dynamic MR imaging findings are correlated with angiogenesis in various tumor types, including cervical carcinoma. In a histologic study (11), all cervical carcinomas showed marked angiogenesis. A correlation between angiogenesis and recurrence was reported by Dinh et al (8), who found that a high microvessel count and tumor size were independent factors predictive of recurrent cervical carcinoma. By following the suggestions raised by these results, one could hypothesize that tumors with a higher risk of recurrence—that is, tumors that demonstrate more aggressive biologic behavior—show a higher degree of angiogenesis, which could potentially be depicted at first-pass contrast-enhanced dynamic MR imaging. Tumors with a high risk of recurrence should then show a high amplitude and a steep slope value during the first-pass of contrast enhancement.

Our results, however, showed no such relationship between biologic aggressiveness of cervical carcinoma and dynamic parameters of contrast-enhanced MR imaging. The dynamic parameters of slope and amplitude did not correlate with the histologic parameters of tumor invasion depth or presence of pelvic lymph node metastases, both of which are strong predictors of tumor recurrence and survival in patients with cancer of the uterine cervix and are established predictors of tumor aggressiveness (5,6,12,13). This corroborates the findings of Wiggins et al (11), who, in a histologic study with 29 patients, found no correlation between microvessel count (a measure of tumor angiogenesis) and nodal status, parametrial involvement, or depth of invasion.

One possible explanation for this apparent discrepancy may be related to insufficient accuracy of the MR imaging method used for reflecting the extent of angiogenesis. As is apparent from the Figure, however, there was a significant difference in slope and amplitude values between tumor and muscle, which indicates that our MR imaging method was, in fact, sensitive enough to demonstrate angiogenesis. One cannot exclude the theoretic possibility, however, that our MR imaging method could not accurately demonstrate true differences in angiogenesis between the various cases of cervical carcinoma. Hawighorst et al (7) have shown such accuracy for MR imaging. The patient population in the study by Hawighorst et al was less homogeneous than the population in our study, however, because there was a larger proportion of patients with more advanced disease (71% had a FIGO stage IIB–IVA tumor).

It is possible that the association between angiogenesis and dynamic contrast-enhancement MR imaging findings is not simple but rather is complex. Angiogenesis, on the one hand, may be defined as the division rate of vascular cells or the spatial density of capillaries. Time–signal intensity curves measured with MR images, on the other hand, represent an interaction between specific blood flow, heterogeneity of flow, diffusion constants in the assessed tissues, spatial variability of these constants, and volumes of distribution of the enhancing agent. A more in-depth review of this topic has been provided by Verstraete et al (10).

Another tentative explanation might be that the hypothesis of a correlation between angiogenesis and the risk of tumor recurrence does not hold true, at least not in the individual patient who is generally considered for radical surgical resection. Dinh et al (8) indeed found a correlation between angiogenesis and recurrence, but on closer inspection, the reported significance of this correlation may be due to the presence of a single outlier; in addition, their study may lack statistical power because of the small number of subjects (n = 22). These observations cast doubt on the purported correlation between angiogenesis and recurrence.

Our conclusion, therefore, is that although dynamic contrast-enhanced MR imaging results reflect the degree of angiogenesis, they cannot be used to determine clinical treatment in the individual patient. In patients who are candidates for radical hysterectomy, one cannot use the dynamic parameters of contrast-enhancement slope and amplitude as tools to help predict tumor aggressiveness in cases of cervical carcinoma.


   Footnotes
 
Current address: Department of Radiology, Erasmus University Medical Center Rotterdam, the Netherlands. Back

Address reprint requests to S.P., Cardington House, Sunning Ave, Sunningdale, Berkshire, SL5 9PW England.

Abbreviation: FIGO = International Federation of Gynecology and Obstetrics

Author contributions: Guarantor of integrity of entire study, S.P.; study concepts, S.P., P.M.T.P.; study design, S.P., P.M.T.P., J.B.T.; definition of intellectual content, S.P.; literature research, S.P., C.S.P.v.R.; clinical studies, J.B.T.; data acquisition, C.S.P.v.R., J.B.T., S.P., P.M.T.P.; data analysis, S.P., P.M.T.P.; statistical analysis, S.P., P.M.T.P.; manuscript preparation, S.P., P.M.T.P.; manuscript editing, C.S.P.v.R., S.P., P.M.T.P.; manuscript review, S.P., P.M.T.P., J.B.T.

Received April 7, 1998; revision requested May 19, 1998; revision received June 12, 1998; accepted August 10, 1998.
   References
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Hricak H, Powell CB, Yu KK, et al. Invasive cervical carcinoma: role of MR imaging in pretreatment work-up—cost minimization and diagnostic efficacy analysis. Radiology 1996; 198:403-409.[Abstract/Free Full Text]
  2. Hricak H, Lacey CG, Sandles LG, Chang YCF, Winkler ML, Stern JL. Invasive cervical carcinoma: comparison of MR imaging and surgical findings. Radiology 1988; 166:623-631.[Abstract/Free Full Text]
  3. Togashi K, Nishimura K, Sagoh T, et al. Carcinoma of the cervix: staging with MR imaging. Radiology 1989; 171:245-251.[Abstract/Free Full Text]
  4. Sironi S, DeCobelli F, Scarfone G, et al. Carcinoma of the cervix: value of plain and gadolinium-enhanced MR imaging in assessing degree of invasiveness. Radiology 1993; 188:797-801.[Abstract/Free Full Text]
  5. Hatch KD. Cervical cancer. In: Berek JS, Hacker NF, eds. Practical gynecologic oncology. 2nd ed. Baltimore, Md: Williams & Wilkins, 1994; 259-261.
  6. Stehman FB, Perez CA, Kurman RJ, Thigpen JT. Uterine cervix. In: Hoskins WJ, Perez CA, Young RC, eds. Principles and practice of gynecologic oncology. 2nd ed. Philadelphia, Pa: Lippincott-Raven, 1997; 819-821.
  7. Hawighorst H, Knapstein PG, Weikel W. Angiogenesis of uterine cervical cancer: characterization by pharmacokinetic magnetic resonance parameters and histological microvessel density with correlation to lymphatic involvement. Cancer Res 1997; 57:4777-4786.[Abstract/Free Full Text]
  8. Dinh TV, Hannigan EV, Smith ER, Hove MJ, Chopra V, To T. Tumor angiogenesis as a predictor of recurrence in stage IB squamous cell carcinoma of the cervix. Obstet Gynecol 1996; 87:751-754.[Abstract]
  9. Hawighorst H, Knapstein PG, Weikel W, Knopp MV, Essig M, Van Kaick G. Dynamic, gadolinium-enhanced MR imaging: a novel prognostic factor to predict lymphatic spread and tumor proliferation in cervical carcinoma (abstr). MAGMA 1996; 4(suppl):83-84.
  10. Verstraete KL, Van der Wouden HJ, Hogendoorn PCW, De Deene Y, Kunnen M, Bloem JL. Dynamic contrast-enhanced MR imaging of musculoskeletal tumors: basic principles and clinical applications. JMRI 1996; 6:311-321.
  11. Wiggins DL, Granai CO, Steinhoff MM, Calabresi P. Tumor angiogenesis as a prognostic factor in cervical carcinoma. Gynecol Oncol 1995; 56:353-356.[Medline]
  12. Delgado G, Bundy BN, Fowler Jr WC. A prospective surgical pathological study of stage I squamous carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 1989; 35:314-320.[Medline]
  13. Delgado G, Bundy B, Zaino R, Sevin BU, Creasman WT, Major F. 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 1990; 38:352-357.[Medline]



This article has been cited by other articles:


Home page
 Am. J. Roentgenol.Home page
A. S. Doria, M. Noseworthy, W. Oakden, R. Moineddin, T. Rayner, V. Tassos, D. Engelberts, K. Pritzker, M. Rogers, R. Jong, et al.
Dynamic contrast-enhanced MRI quantification of synovium microcirculation in experimental arthritis.
Am. J. Roentgenol., April 1, 2006; 186(4): 1165 - 1171.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
J. F. Schaefer, J. Vollmar, F. Schick, R. Vonthein, M. D. Seemann, H. Aebert, R. Dierkesmann, G. Friedel, and C. D. Claussen
Solitary Pulmonary Nodules: Dynamic Contrast-enhanced MR Imaging--Perfusion Differences in Malignant and Benign Lesions
Radiology, August 1, 2004; 232(2): 544 - 553.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
H. Hawighorst, S. Postema, and P. M. T. Pattynama
Dynamic MR Imaging in Cervical Carcinoma • Drs Postema and Pattynama respond:
Radiology, October 1, 1999; 213(2): 617 - 618.
[Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Postema, S.
Right arrow Articles by Trimbos, J. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Postema, S.
Right arrow Articles by Trimbos, J. B.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE