The use of combined PET/CT for localizing recurrent head and neck cancer: the Pittsburgh experience.

Abstract

We performed a retrospective study of 47 patients to ascertain the ability of combined positron-emission tomography and computed tomography (PET/CT) to localize recurrent head and neck cancer. When clinically warranted, biopsies were performed in an attempt to obtain pathologic confirmation of the PET/CT findings. Of the 47 patients, 33 exhibited PET/CT findings consistent with recurrent cancer. Of the 33 patients, 25 underwent either biopsy or surgical excision of disease in an attempt to obtain a pathologic confirmation. Biopsy analysis confirmed the PET/CT findings in 22 of these patients; in the remaining 3 patients, pathologic findings were inconsistent with the PET/CT diagnosis. Based on the subset of 25 patients who underwent pathologic testing, the sensitivity of combined PET/CT was 95% and the specificity was 60%. We conclude that combined PET/CT imaging is a valuable tool for localizing tumor recurrence in patients with head and neck cancer.

Introduction

The diagnosis of recurrent head and neck cancer is a challenge in patients treated with radiation therapy, surgery, or both. Postsurgical anatomic changes and inflammation associated with radiation limit the diagnostic accuracy of traditional imaging modalities such as computed tomography (CT) and magnetic resonance imaging (MRI). (1) Alternatively, the use of panendoscopy with biopsy of irradiated tissues in patients with a presumed recurrence may promote radionecrosis; furthermore, many patients without recurrence would be exposed to additional surgical and anesthetic risks. An imaging modality that would help localize recurrence and guide the pathologic diagnosis of cancer in these patients would be beneficial.

Positron-emission tomography (PET) has been shown to be effective in detecting a variety of cancers in different areas of the head and neck. (2,3) Accelerated glucose metabolism, reflected by a focally increased uptake of [18F]2-fluoro-2-deoxyglucose (FDG), is highly correlated with malignancy. However, the limited spatial resolution and the lack of anatomic detail associated with PET often make it difficult to precisely localize a tumor, given the variable degree of physiologic FDG uptake that occurs in the head and neck. This inability to precisely localize a tumor can confound diagnosis, staging, and treatment planning.

A prototype combination PET/CT scanner was developed at the University of Pittsburgh to localize neoplastic lesions throughout the body. (4,5) Initial studies demonstrated that PET/CT was more effective than PET alone in this regard. In this article, we describe our early experience with combined PET/CT in patients with recurrent head and neck cancer.

Patients and methods

We examined the records of 47 patients--29 men and 18 women, aged 51 to 102 years (mean: 64.1 [+ or -] 18.2)--who had been referred by the Department of Otolaryngology-Head and Neck Surgery to the Department of Radiology at the University of Pittsburgh for PET/CT evaluation of suspected head and neck cancer between Oct. 1, 2000, and June 30, 2001. The use of the investigational prototype PET/CT scanner for this purpose was approved by our institutional review board, and all patients signed consent forms approved by the board.

Imaging equipment. The prototype PET/CT scanner that we used combines elements of the AR.SP CT scanner (Siemens Medical Systems; Erlanger, Germany) and the ECAT ART PET scanner (CTI PET Systems; Knoxville, Tenn.). The PET and CT components were mounted on the same assembly and offset axially by 60 cm. A moving bed allowed for dual-modality imaging with an axial extent of 100 cm. Data were acquired sequentially; the CT images were obtained first. Detailed technical information on this equipment can be found in previous publications. (4,5)

Patients were injected with approximately 260 MBq of FDG. After an interval of approximately 60 minutes to allow for adequate FDG uptake, patients were positioned in the scanner. A scout CT was performed to determine the axial range of the image. A helical CT of the neck through the chest was performed during shallow breathing. The total CT scanning time was approximately 5 minutes. Immediately following CT, PET data were acquired (10 min per bed position). The total scan time was approximately 45 minutes.

Image analysis. Each image was reviewed by 2 senior radiologists, 1 of whom was board-certified in nuclear medicine. Lesions were visually characterized according to their location and the likelihood that they were malignant, based on FDG uptake. Imaging findings were retrospectively compared with clinical information, including CT reports contained in the patient records when available.

Surgery. Patients with positive findings on combined PET/CT were evaluated for surgery by senior staff otolaryngologists. Surgical specimens were sent for routine pathologic processing, and senior staff pathologists evaluated and diagnosed each specimen. Patients with negative findings on PET/CT were followed clinically by staff otolaryngologists.

Results

Of the 47 patients, 29 had undergone a previous surgical resection of head and neck cancer with or without chemoradiation, and 18 had undergone chemoradiation only. A total of 31 patients had biopsy-proven squamous cell carcinoma of the head and neck; the remaining 16 had various other types of tumors.

Of the 47 patients, 33 exhibited PET/CT findings consistent with cancer. Of this group, 25 underwent either biopsy or surgical excision of the tumor to obtain material for a pathologic diagnosis; the remaining 8 patients were referred to the oncology unit for treatment without pathologic confirmation. Three of these 8 refused further biopsies or surgery, and pathologic testing prior to medical treatment was not pursued in the other 5. The 14 patients whose imaging results were negative were followed with routine physical examinations and repeat PET/CT evaluations every 3 to 6 months.

In the 25 patients who underwent pathologic testing, the average length of time between the original surgical removal of disease and the PET/CT study was 22 months ([+ or -] 10). In this group, pathology confirmed the PET/CT findings in 22 patients. There were 19 true positives and 3 true negatives (table). Of the 3 patients whose pathologic findings were inconsistent with PET/CT, there were 2 false positives and 1 false negative. Therefore, the sensitivity of PET/CT was 95%, the specificity was 60%, and the positive predictive value (PPV) was 90%.

The location of recurrence varied; 19 of the 25 patients (76%) demonstrated local recurrence, 10 (40%) had lymphatic recurrence, and 3 (12%) developed distant metastatic disease (table).

Only a few patients underwent concurrent radiologic examinations by other modalities (figures 1, 2, and 3) and, therefore, a definitive direct comparison of PET/CT with CT alone and with MRI could not be addressed in this study.

[FIGURES 1-3 OMITTED]

Eighteen of the 25 patients (72%) had squamous cell carcinoma. The sensitivity and specificity of PET/CT in patients with recurrent squamous cell carcinoma were 94 and 50%, respectively, and the PPV was 94%. The sensitivity and specificity of PET/CT in the remaining 7 patients were 100 and 67% respectively, and the PPV was 71%.

Discussion

Our data indicate that combined PET/CT is a useful tool for identifying and precisely localizing recurrent tumors in the head and neck. This precise anatomic localization may allow for targeted surgical excision and/or radiotherapy (brachytherapy). Furthermore, some patients with advanced disease may be referred directly for chemotherapy based on PET/CT results and therefore would avoid the morbidity associated with surgery.

In our study, PET/CT yielded an overall sensitivity of 95% and a specificity of 60% for identifying recurrent disease in the head and neck in patients undergoing pathologic testing for confirmation. The low specificity is likely attributable to the fact that patients with negative imaging findings did not undergo testing for pathologic confirmation. Nevertheless, the sensitivity and specificity we found are in close agreement with those of previous studies of PET alone in the detection of head and neck cancers. (6,7)

Another important role of PET/CT may be the identification of recurrent disease in postoperative and post-irradiated tissues. Altered tissue planes, scarring, and radiation-induced necrosis and edema limit the usefulness of traditional imaging techniques. Moreover, physical examination and conventional imaging techniques have been limited, as well.

In previous studies, PET alone identified recurrences in posttreatment necks with 100% accuracy, whereas the accuracy of CT and physical examination was only 53 and 63%, respectively. (8) In 1994, Greven et al reported that PET accurately identified recurrences in 5 of 5 patients with laryngeal cancer. (9) In 1999, Hanasono et al reported that PET yielded a specificity and sensitivity of 100 and 85%, respectively, for recurrent or residual disease in head and neck cancer patients. (10) The individual sensitivity of CT and MRI was 100 and 50%, respectively, and the specificity was 75 and 50%.

However, it is well known that a major limitation of PET alone is its inability to localize the precise anatomic site of recurrence. Studies of retrospective registration of PET and CT or PET and MRI data showed that the availability of both anatomic and functional information allowed for the identification of more tumor sites and tumor margins than did conventional imaging modalities. (8) Although we did not directly compare the use of PET/CT with either CT alone or with MRI, other authors have. (6,7) They found that PET/CT can be useful in evaluating patients whose CT and/or MRI results are equivocal.

The timing of PET following radiotherapy, and possibly following surgery, can affect its utility in tumor surveillance. In 1997, Keyes et al reported that PET yielded a high false-negative rate of 17% (7/41) at 1 month following radiotherapy; at 4 and 12 months, however, there were no false negatives. (11) Even so, early PET is still superior to both physical examination and conventional imaging in this patient population. (12)

PET/CT is not without its own limitations. The PET component is limited by its poor spatial resolution. Moreover, PET is associated with a low sensitivity for tumors smaller than 1 cm, which can hinder the identification of microdeposits of tumor. (10,13,14) Limited resolution and high background counts in secretions may limit the detection of small lesions of the tongue base, tonsils, and salivary glands. Newer generations of PET/CT scanners and the use of scanning agents that are not secreted in saliva may further improve the detection rate of cancer in the head and neck.

Our study was limited by an inherent referral bias, as only patients with a clinical suspicion with or without CT and/or MRI were referred for PET/CT. Likewise, our study was limited by an evaluation bias, as patients with negative PET/CT scans and a low index of suspicion did not undergo testing for pathologic confirmation, which precluded us from interpreting negative PET/CT studies. Prospective, randomized studies are warranted to compare the efficacy of PET/CT with PET alone and CT alone.

In conclusion, our results indicate that combined PET/CT is a valuable tool for the surveillance of recurrent and/or residual head and neck cancer following surgery and/or chemoradiation. The information it provides can allow for the early detection of disease that is not evident on traditional imaging modalities, and it can allow for a greater chance at cure or palliation with less surgical morbidity.

Table. Primary pathologic diagnosis, location of the primary tumor,
PET/CT result, and the location of recurrence in patients who
underwent pathologic testing for confirmation of PET/CT findings

                     Location of the     PET/CT
Pt.  Pathology       primary tumor       result

  1  SCC             Tongue base           TN
  2  SCC             Larynx                FN
  3  SCC             Tongue base           TP
  4  SCC             Tongue base           TP
  5  SCC             Palate                TP

  6  ASC             Floor of the mouth    TP
  7  BCC             Temporal bone         TP
  8  Osteosarcoma    Maxillary sinus       TP
  9  Adenocarcinoma  Palate                TP
 10  SCC             Larynx                TP

 11  SCC             Tongue base           TP
 12  Adenocarcinoma  Tongue base           TN
 13  SCC             Tongue base           TP
 14  SCC             Tongue base           TP
 15  SCC             Tongue base           TP

 16  SCC             Buccal mucosa         TP
 17  SCC             Esophagus             TP
 18  SCC             Ethmoid sinus         TP
 19  SCC             Larynx                TP
 20  SCC             Larynx                FP

 21  CEA             Parotid               TN
 22  SCC             Supraglottis          TP
 23  SCC             Tongue                TP
 24  Lymphoma        Tongue                FP
 25  SCC             Larynx                TP

                         Site of recurrence

Pt.  Pathology       Local  Lymphatics  Distant

  1  SCC               +        -          -
  2  SCC               +        -          -
  3  SCC               +        +          -
  4  SCC               +        -          -
  5  SCC               +        +          -

  6  ASC               +        -          -
  7  BCC               +        +          +
  8  Osteosarcoma      +        +          -
  9  Adenocarcinoma    +        -          -
 10  SCC               +        +          +

 11  SCC               +        -          -
 12  Adenocarcinoma    -        -          -
 13  SCC               +        +          -
 14  SCC               +        -          -
 15  SCC               +        -          -

 16  SCC               -        +          -
 17  SCC               +        -          -
 18  SCC               +        -          -
 19  SCC               +        +          -
 20  SCC               -        -          -

 21  CEA               -        -          -
 22  SCC               +        +          -
 23  SCC               +        +          -
 24  Lymphoma          -        -          -
 25  SCC               -        -          +

ASC = adenosquamous carcinoma; BCC = basal cell carcinoma; CEA =
carcinoma ex pleomorphic adenoma; FN = false negative; FP = false
positive; Pt. = patient; SCC = squamous cell carcinoma; TN = true
negative; TP = true positive.

References

(1.) Glazer HS, Niemeyer JH, Balfe DM, et al. Neck neoplasms: MR imaging. Part II. Posttreatment evaluation. Radiology 1986;160:349-54.

(2.) Jabour BA, Choi Y, Hoh CK, et al. Extracranial head and neck: PET imaging with 2-[F-18]fluoro-2-deoxy-D-glucose and MR imaging correlation. Radiology 1993;186:27-35.

(3.) Bailet JW, Abemayor E, Jabour BA. et al. Positron emission tomography: A new, precise imaging modality for detection of primary head and neck tumors and assessment of cervical adenopathy. Laryngoscope 1992:102:281-8.

(4.) 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;25:905-10.

(5.) Beyer T, Townsend DW, Brun T, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med 2000;41:1369-79.

(6.) Di Martino E. Nowak B, Hassan HA, et al. Diagnosis and staging of head and neck cancer: A comparison of modern imaging modalities (positron emission tomography, computed tomography, color-coded duplex sonography) with panendoscopic and histopathologic findings. Arch Otolaryngol Head Neck Surg 2000;126:1457-61.

(7.) Lonneux M, Lawson G, Ide C, et al. Positron emission tomography with fluorodeoxyglucose for suspected head and neck tumor recurrence in the symptomatic patient. Laryngoscope 2000;110:1493-7.

(8.) Wong WL, Hussain K, Chevretton E, et al. Validation and clinical application of computer-combined computed tomography and positron emission tomography with 2-[18F]fluoro-2-deoxy-D-glucose head and neck images. Am J Surg 1996;172:628-32.

(9.) Greven KM, Williams DW III, Keyes JW, Jr., et al. Distinguishing tumor recurrence from irradiation sequelae with positron emission tomography in patients treated for larynx cancer. Int J Radiat Oncol Biol Phys 1994;29:841-5.

(10.) Hanasono MM, Kunda LD, Segall GM, et al. Uses and limitations of FDG positron emission tomography in patients with head and neck cancer. Laryngoscope 1999;109:880-5.

(11.) Keyes JW, Jr., Watson NE, Jr., Williams DW III, et al. FDG PET in head and neck cancer. AJR Am J Roentgenol 1997;169:1663-9.

(12.) Lowe VJ, Dunphy FR, Varvares M, et al. Evaluation of chemotherapy response in patients with advanced head and neck cancer using [F-18]fluorodeoxyglucose positron emission tomography. Head Neck 1997;19:666-74.

(13.) McGuirt WF, Greven K, Williams D III, et al. PET scanning in head and neck oncology: A review. Head Neck 1998;20:208-15.

(14.) Kau RJ, Alexiou C, Laubenbacher C, et al. Lymph node detection of head and neck squamous cell carcinomas by positron emission tomography with fluorodeoxyglucose F 18 in a routine clinical setting. Arch Otolaryngol Head Neck Surg 1999;125:1322-8.

From the Department of Otolaryngology Head and Neck Surgery (Dr. Zimmer and Dr. Snyderman), the Department of Radiology (Dr. Fukui, Dr. Blodgett, Dr. McCook, Dr. Townsend, and Dr. Meltzer), and the Department of Psychiatry and the Department of Neurology (Dr. Meltzer), University of Pittsburgh Medical Center.

Reprint requests: Lee A. Zimmer, MD, Department of Otolaryngology, University of Pittsburgh Eye and Ear Institute, 200 Lothrop St., Pittsburgh, PA 15213-2546. Phone: [phone]: fax: [phone]; e-mail: [e-mail]

Originally presented at the Sixth Workshop on the Biology, Prevention, and Treatment of Head and Neck Cancer; Oct. 1, 2002; Washington, D.C.