Todd Blodgett, MD

The American Cancer Society 1 has estimated that, in 2003, there will be 25,400 cases of newly diagnosed ovarian cancer in the United States, with 14,300 deaths; 12,200 cases of cervical cancer, with 4,100 deaths; and 40,100 cases of uterine cancer, with 6,800 deaths. This accounts for a significant portion of morbidity and mortality in the female population. Clinical examination, combined with noninvasive anatomical imaging modalities such as ultrasound, CT, and MRI, has been the most commonly employed method used to stage and restage the conditions of these cancer patients. 2-5 Some studies, 2,3,6 however, have shown that CT and MRI have an unacceptably low sensitivity and specificity in this patient population.

Early experience in the use of whole-body dedicated positron-emission tomography (PET) imaging for gynecological malignancies has indicated a similar or slightly higher sensitivity, specificity, and accuracy (compared with CT and MRI) in cancer staging, as well as in the detection of recurrent disease. 7-11 The limitation of dedicated PET is the lack of correlative anatomical data, since physiological uptake of fluorodeoxyglucose (FDG) in the bowel, ureters, and bladder can often confound interpretation of PET in the abdomen and pelvis, a complex anatomical region. In addition, PET scanners have an absolute limitation in terms of spatial resolution, currently in the range of 4 to 8 mm, depending on the scanner and the metabolic activity of the lesion. Furthermore, tracer concentration in lesions of less than 1 cm may be underestimated due to partial-volume averaging; thus, small areas or nodes within the mesentery may be misinterpreted as areas of physiological uptake, in the absence of accurate correlative anatomical data.

David Townsend, PhD

Recently, focus has shifted from comparing the sensitivity and specificity of anatomical and metabolic imaging modalities to the development and implementation of a combined anatomical and functional scanner that has the ability to fuse the CT and PET data sets accurately by acquiring both modalities in the same scanner. 12,13 Visual comparison of separately acquired PET and CT data and retrospective registration using a variety of software approaches, although sometimes helpful, are generally cumbersome and tend to work less effectively in the abdomen and pelvis, where there is significant potential for organ movement between the two scans. In addition, many of the PET findings that would actually benefit from accurate coregistration correspond to subtle anatomical abnormalities, so the interpreting physician relies on the accuracy of the actual registration process to make a confident, definitive diagnosis. In fact, in some cases, there is no correlative anatomical abnormality at all. In these cases, the combined PET/CT technique is superior because it allows the interpreting physician to have the confidence to convey the location of the abnormality accurately to a surgeon or referring clinician. Without a correlative anatomical abnormality, the fused images rendered by retrospective registration programs need to be evaluated with considerable caution.

Figure 1: Ovarian cancer recurrence. (A,B) Initial axial CT and PET/CT images demonstrate a hypermetabolic malignant focus in this patient who had a rising CA-125 level and a negative CT scan prior to PET/CT evaluation. The abnormality (arrow) localized to the peripancreatic area by use of the fused image is not well seen on the CT portion of the exam. The patient was sent for a CT-guided biopsy, but the lesion could not be located and the patient was lost to follow-up at that time. Subsequent axial CT and PET/CT images (C,D) performed more than a year later shows diffuse malignancy throughout the abdomen.

At the University of Pittsburgh, we have had the opportunity, over the past 5 years, to scan more than 3,000 patients using a prototype PET/CT scanner, 12,13 two commercial PET/CT scanners, a bismuth germanium oxide system, and a lutetium oxyorthosilicate scanner. We have identified a number of useful clinical applications of PET/CT in gynecological malignancies. One of the major factors limiting the use of PET/CT in gynecological malignancies, however, is the current lack of insurance reimbursement for PET use for these indications in the United States. Although there have been some encouraging studies 14-16 using PET for some of these applications, particularly in patients with suspected ovarian cancer recurrence, most data are inconclusive; some studies 17-19 demonstrated disappointing results for PET.

Approximately 200 patients with ovarian, cervical, or endometrial cancer, in different stages of the disease process, have been scanned at the University of Pittsburgh Medical Center. Although we have not performed a definitive analysis of these data, preliminary results are encouraging. Of the initial 26 patients with primary and recurrent gynecological malignancies (15 ovarian and 11 cervical) who were evaluated for treatment planning at the University of Pittsburgh Medical Center, PET/CT identified additional lesions not noted on their traditional pretreatment clinical CT studies in five of 11 (45%) cervical-cancer patients and 12 of 15 (80%) ovarian-cancer patients. The actual significance of these findings is the effect that the identification of the additional lesions had on patient management, which changed in 17 of 26 (65%) patients. Three patients’ treatments were changed from possible surgery of an isolated lesion to chemotherapy due to the identification of remote additional lesions. Ten patients’ treatments were changed from observation (due to negative or equivocal CT scans) to salvage chemotherapy, and four patients had extension of radiation fields to cover additional lesions. 20,21

Figure 2. Radiation-therapy planning and cervical cancer: a staging clinical CT scan showed a cervical mass and para-aortic lymphadenopathy. A pretreatment coronal positron-emission tomography (PET)-CT image (A) showed additional para-aortic involvement and a small, hypermetabolic left pelvic-sidewall lymph node (B and C). This changed the radiation plan by adding 12 mm of additional para-aortic coverage and intensity-modulated radiation therapy to cover the abnormal lymph node. Post-treatment coronal PET/CT (D) showed a significant response to radiation therapy.

Three cases demonstrate some of the potentially useful clinical applications of the combined modality in patients with gynecological malignancies. Figure 1 comes from a patient who was referred for suspected recurrence of ovarian cancer due to an increase in her cancer antigen 125 level. A CT performed at an external institution just prior to her PET/CT study was interpreted as normal, even in retrospect. After an abnormality in the peripancreatic area had been identified using PET/CT, this patient was sent for a CT-guided biopsy of the lesion. When the patient was on the biopsy table, however, the lesion could not be identified. Attempts were made to provide the fused PET/CT images to the interventional suite by courier to help guide the biopsy, but the patient refused the examination and was lost to follow-up care. She returned over a year later for a repeat PET/CT study; it showed diffusely progressive metastatic disease, including abdominal lymphadenopathy and multiple masses within the mesentery. We now routinely send fused images with patients who are going to have surgery or biopsies performed, particularly when we feel that the fused image would be helpful in guiding the physician performing the procedure.

We have also been working closely with radiation oncologists to define tumor extent and target volumes more accurately. Figure 2 shows a patient with recently diagnosed cervical cancer who was referred for staging prior to resection and radiation therapy. Compared with the pretreatment staging CT, PET/CT was able to define an additional 12 mm of abnormal para-aortic nodal involvement along the craniocaudal extent, as well as identifying a small abnormal left-sided pelvic lymph node (for which additional intensity-modulated radiation therapy was prescribed).

Figure 3. Small ovarian-cancer recurrence in two patients: one patient had a negative CT scan prior to having qxial CT and PET/CT performed; resulting images (A and B) demonstrate a focal hypermetabolic mesenteric implant of 7 to 8 mm adjacent to a loop of bowel (arrow). A second patient had a subtle hypermetabolic focus identified and localized to the abdominal wall, as demonstrated by axial CT and PET/CT (C and D).

Although not all patients with gynecological malignancies are being sent for pretreatment PET/CT scanning, patients are increasingly being referred for this reason. Our experience is that PET/CT offers a much more accurate definition of the extent of disease and target volumes for radiation therapy planning in this patient population, although more studies are needed to confirm these findings.

Because bowel can demonstrate both focal and diffuse patterns of physiologic FDG uptake, it can be a diagnostic challenge to identify small areas of tumor involvement in the abdomen and pelvis, particularly when lesions are adjacent to normal bowel. Figure 3 demonstrates this difficulty in PET imaging and shows that, with a combined PET/CT scanner, these lesions are much more easily identified.

Although more studies are needed to define the multiple potential roles of combined PET/CT in managing gynecological malignancies, we are seeing a significant increase in the number of patients sent for evaluation due to the success we have demonstrated with the patients scanned to date. Of course, there are still challenges for PET/CT, as it obviously retains some of the individual limitations of separate PET and CT scanning. For example, neither PET nor CT is yet capable of detecting microscopic disease, PET because of spatial-resolution limitations and CT because of the absence of an identifiable anatomical correlate. In addition, there will continue to be false-positive studies due to the identification of hypermetabolic nodes that are reactive, but do not actually contain tumor cells. In our experience, these cases are the exception, and the value gained from the modality far outweighs any potential limitations.

The future of PET imaging lies in the development of more specific imaging probes; as this progresses, combined anatomical and functional imaging will become imperative due to the lack of uptake in background tissue and the increasing absence of anatomical detail. Regardless of the specificity of the probes developed, it is becoming evident that combined PET/CT imaging will continue to have an increasingly significant impact on oncology.

Todd Blodgett, MD, is fellowship-trained in PET and currently completing his radiology residency in the research track at the University of Pittsburgh Department of Radiology.

David Townsend, PhD, is Professor and Director, Cancer Imaging and Tracer Development Program, Department of Medicine, University of Tennessee, Knoxville.

References:

  1. American Cancer Society. Cancer facts & figures 2002. Available at: http://www.cancer.org/docroot/STT/content/STT_1x_Cancer_Facts__Figures_2002.asp. Accessed June 12, 2003.
  2. Megibow AJ. Accuracy of CT in detection of persistent or recurrent ovarian carcinoma: correlation with second-look laparotomy. Radiology. 1988;166:341-345.
  3. Prayer L. CT and MR accuracy in the detection of tumor recurrence in patients treated for ovarian cancer. J Comput Assist Tomogr. 1993;17:626-632.
  4. Low RN, Saleh F, Song SY, et al. Treated ovarian cancer: comparison of MR imaging with serum CA-125 level and physical examination–a longitudinal study. Radiology. 1999;211:519-528.
  5. Nelson BE, Rosenfield AT, Schwartz PE. Preoperative abdominopelvic computed tomographic prediction of optimal cytoreduction in epithelial ovarian carcinoma. J Clin Oncol. 1993;11:166-172.
  6. Goldhirsch A, Triller JK, Greiner R, Dreher E, Davis BW. Computed tomography prior to second-look operation in advanced ovarian cancer. Obstet Gynecol. 1983;62:630-634.
  7. Kubik-Huch RA, Dorffler W, von Schulthess GK, et al. Value of (18F)-FDG positron emission tomography, computed tomography, and magnetic resonance imaging in diagnosing primary and recurrent ovarian carcinoma. Eur Radiol. 2000;10:761-767.
  8. Zimny M, Siggelkow W, Schroder W, et al. 2-[Fluorine-18]-fluoro-2-deoxy-d-glucose positron emission tomography in the diagnosis of recurrent ovarian cancer. Gynecol Oncol. 2001;83:310-315.
  9. Hubner KF, McDonald TW, Niethammer JG, Smith GT, Gould HR, Buonocore E. Assessment of primary and metastatic ovarian cancer by positron emission tomography (PET) using 2-[18F]deoxyglucose (2-[18F]FDG). Gynecol Oncol. 1993;51:197-204.
  10. Karlan BY, Hoh C, Tse N, Futoran R, Hawkins R, Glaspy J. Whole-body positron emission tomography with (fluorine-18)-2-deoxyglucose can detect metastatic carcinoma of the fallopian tube. Gynecol Oncol. 1993;49:383-388.
  11. Yuan CC, Liu RS, Wang PH, Ng HT, Yeh SH. Whole-body PET with (fluorine-18)-2-deoxyglucose for detecting recurrent ovarian carcinoma. Initial report. J Reprod Med. 1999;44:775-778.
  12. Beyer T, Townsend DW, Brun T, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med. 2000;41:1369-1379.
  13. Charron M, Beyer T, Bohnen NN, et al. Image analysis in patients with cancer studied with a combined PET and CT scanner. Clin Nucl Med. 2000;25:905-910.
  14. Chang WC, Hung YC, Kao CH, Yen, RF, Shen YY, Lin CC. Usefulness of whole body positron emission tomography (PET) with 18F-fluoro-2-deoxyglucose (FDG) to detect ovarian cancer based on asymptomatically elevated serum levels of tumor marker. Neoplasma. 2002;49(5):329-33.
  15. Torizuka T, Nobezawa S, Kanno T, et al. Ovarian cancer recurrence: role of whole-body positron emission tomography using 2-[fluorine-18]-fluoro-2-deoxy- D-glucose. Eur J Nucl Med Mol Imaging. 2002;29(6):797-803.
  16. Bristow RE, Simpkins F, Pannu HK, Fishman EK, Montz FJ. Positron emission tomography for detecting clinically occult surgically resectable metastatic ovarian cancer. Gynecol Oncol. 2002;85(1):196-200.
  17. Cho SM, Ha HK, Byun JY, Lee JM, Kim CJ, Nam-Koong SE. Usefulness of FDG PET for assessment of early recurrent epithelial ovarian cancer. AJR Am J Roentgenol. 2002;179(2):391-5.
  18. Fenchel S, Grab D, Nuessle K, Kotzerke J, Rieber A, Kreienberg R, et al. Asymptomatic adnexal masses: correlation of FDG PET and histopathologic findings. Radiology. 2002;223(3):780-8.
  19. Rose PG, Faulhaber P, Miraldi F, Abdul-Karim FW. Positive emission tomography for evaluating a complete clinical response in patients with ovarian or peritoneal carcinoma: correlation with second-look laparotomy. Gynecol Oncol. 2001;82(1):17-21.
  20. Blodgett TM, Meltzer CC, Townsend DW, et al. PET/CT in re-staging patients with ovarian cancer [abstract]. J Nucl Med. 2002;45:310.
  21. Blodgett TM, Meltzer CC, Townsend DW, et al. PET/CT in staging and re-staging patients with cervical cancer [abstract]. J Nucl Med. 2002;45:310.