Original article

FGD-PET in the follow-up of recurrent colorectal cancer

F. Selvaggi*, A. Cuocolo†, G. Sciaudone*, S. Maurea†, A. Giuliani* and C. Mainolfi†

*1st Division of Genral Surgery, Second University of Naples and †Service of Nuclear Medicine, University of Naples ‘Federico II’, Naples, Italy

Received 3 October 2002; accepted 1 December 2002

Abstract

Objective The current methods of detection of recur-

rent colorectal cancer after surgical treatment are inac-

curate using conventional imaging. This study set out to

detect early recurrence by means of PET in patients

treated surgically for colorectal cancer by curative resec-

tion.

Methods Thirty-one disease-free patients were recrui-

ted and underwent FDG-PET. The results were verified

by clinical, surgical and radiological follow up and ⁄ or

biopsy to evaluate the accuracy for detecting recurrence.

Results PET detected 6 sites of increased activity in 5

patients. Three of these underwent surgery. One was false

positive with no evident tumour and two underwent an

hepatic resection with removal of a homental metastasis.

The sensitivity was 100% and specificity 83.3%. Clinical

management was altered in two cases (6.4%).

Conclusions This study demostrates that PET is more

accurate than conventional imaging for the evaluation of

recurrence in colorectal cancer patients. FDG-PET

should be considered in the follow-up of patients after

treatment for colorectal cancer in addition to other

imaging methods.

Keywords Colorectal cancer, PET, recurence, follow-up

Introduction

Recurrence after surgical treatment of colorectal cancer

(CRC) develops in 25% to 50% of patients, usually within

two years of the first resection [1]. Current methods for

the detection of recurrence include clinical examination,

computed tomography (CT) [2], magnetic resonance

imaging (MRI) [3], increase of carcinoembryonic antigen

(CEA) serum levels [4] and more recently metabolic

imaging with fluorodeoxyglucose (FDG) positron emis-

sion tomography (PET) [5]. The major goal of follow-up

strategies is to diagnose recurrent disease as early as

possible and to select those patients who might benefit

from re-operation or adjuvant therapy [6].

The aim of this study was to assess the role of

metabolic PET imaging with FDG in the detection of

early recurrence in patients treated surgically for CRC

having no evidence of recurrent disease based on clinical

examination, tumour markers (CEA), endoscopy and CT

and MRI scanning.

Methods

Patients

Forty-nine consecutive patients, with an histologically

documented diagnosis of colorectal cancer, underwent

curative surgical resection between March 1993 and

December 1998. Diabetic patients were excluded from

the study because high glucose serum levels competitively

inhibit FDG transport into malignant cells, potentially

decreasing the ability to detect and characterize tumour

[7].

Thirty-one disease-free patients (19 male; 12 female)

with a mean age of 61.6 years (range 43–79 years) were

selected. The site of localization, type of surgical proce-

dure and tumour staging are shown in Tables 1, 2 and 3.

These patients were considered disease free after per-

forming total body CT scan and MRI at one month, one

and two years after surgical treatment.

Study protocol

All patients underwent CT scan at six and 12 months

using a Somaton plus-S scanner (Samiton, Siemens,

Erlangen, Germany) with a dynamic technique taking

10-mm thick sections. MRI was performed using a Vectra

Correspondence to: Dr Francesco Selvaggi, Via Francesco Giordani, 42, 80122

Naples, Italy.

E-mail: fselvaggi@hotmail.com

496 � 2003 Blackwell Publishing Ltd. Colorectal Disease, 5, 496–500

0.5 T machine (General Electric, Milwaukee, WI, USA)

with T1 weighted spin-echo images (repetition time

500 ms; echo time 30 ms) and T2 weighted spin echo

images (repetition time 1800 ms; echo time 80 ms). The

median thickness was 10 mm.

All patients underwent clinical examination (every

three months) and CEA evaluation in addition. Colono-

scopy was performed at one month, one and two years

after surgical treatment. Whole body FDG PET was

performed two years after surgery in all patients.

Flourine-18 FDG PET Imaging

A MC-170d cyclotron (Scanditronix, Uppsala, Sweden)

produced flourine-18 (F-18). F-18 was transferred to an

automatic system for synthesis of F-18 FDG. The quality

of F-18 FDG production was tested for sterility, pyroge-

nicity and radiochemical purity. PET imaging was

performed using a whole-body PET EXACT 47 scanner

(Siemens, Erlangen, Germany). This tomograph has a

16.2 axial field of view and yields 47 image planes for bed

position. Patients were positioned on the PET gantry

using rectilinear scan computerized program localized to

the upper abdomen.

Before injection of F-18 FDG, an abdominal and

pelvic transmission scan was made using a rod source of

Ge-67 for the correction of the corresponding emission

scans. Patients then had an intravenous injection of of

F-18 FDG (370 MBq). Abdomen and pelvis emission

imaging was acquired between 30 and 45 min after FDG

administration. Finally, whole-body images using seven-

bed position with an acquisition time of 6 min for each

position were performed within 1 h after tracer injection.

No brain FDG activity was included.

Images were reconstructed using filtered back projec-

tion smoothed with a Hann filter with a cut off frequency

of 0.4 cycles ⁄ pixels by SUN workstation system (Sie-

mens, Erlangen, Germany) generating three-dimensional

PET scans as axial, coronal and sagittal views [7].

Two independent and experienced nuclear medicine

physicians evaluated the PET-FDG images on a high-

resolution display. In the case of disagreement, a final

interpretation was determined by consensus. The PET

scans were assessed independently without knowledge of

clinical and imaging findings. The presence of abnormal

increased FDG uptake in tumour sites was considered

likely when tracer uptake was greater than blood pool as

well as background activity; in particular a significant

FDG accumulation in tumour was defined as a focus of

increased tracer uptake above the intensity of the

surrounding activity. On whole-body PET images, the

presence of abnormal FDG uptake at tumour sites was

assessed using the same criteria previously mentioned for

the evaluation of abdomen and pelvis PET images [7].

Results

At two years follow-up, 31 patients had no abnormal

clinical findings and normal CEA blood levels. Colonos-

copy and total body CT scan and MRI were negative for

local recurrence or distant metastasis.

PET-FDG imaging detected 6 sites with increase

activity in 5 patients (Table 4) in the liver (n ¼ 3), pelvis

(n ¼ 1), intra-abdominal (n ¼ 1) and in the lung

(n ¼ 1) (Figs 1–4).

In 1 patient, a metastasis in the greater omentum was

discovered and removed in the case with intra-abdominal

by PET. In another with localisation to the pelvis location

at PET no tumour was found at laparotomy (false

positive). The CT scan and PET-FDG two years later

were negative.

A third patient with a positive PET-FDG scan, was

shown, by CT 6 months later, to have hepatic and

pulmonary metastases confirmed by fine needle biopsy

(FNAB). A fourth patient had multiple hepatic metastases

on CT scan one year after a positive PET scan and a fifth

patient showed a single hepatic metastasis on CT scan,

Table 1 Localization at surgical operation (n ¼ 31).

Site of localization No. of patients (%)

Caecum 2 (6.4)

Right colon 4 (12.9)

Hepatic flexure 1 (3.2)

Left colon 9 (29)

Sigmoid 7 (22.5)

Rectum 8 (25.8)

Table 2 Surgical curative operations (n ¼ 31).

Types of surgical procedures No. of patients

Right colon resection 7

Left colon resection 12

Anterior resection 10

Abdominoperineal resection 2

Table 3 Dukes stage (n ¼ 31).

Tumor staging Number of patients

B2 8

C1 13

C2 10

F. Selvaggi et al. FGD-PET in recurrent colorectal cancer

� 2003 Blackwell Publishing Ltd. Colorectal Disease, 5, 496–500 497

6 months after PET, confirmed by FNAB and underwent

surgery. Of the five patients there was, therefore, one false

positive in PET scan.

Discussion

There are many reports of the incidence of recurrence and

the effectiveness of follow-up in detection. Of 818

patients who had undergone curative resection for Dukes

B2 or Dukes C colorectal cancer, 353 (43%) developed

recurrence over 7 years with a median time of recurrence

of 16.7 months [1]. The liver (33%) and the lungs (22%)

were the most frequent distant sites of recurrence [1].

A comparison of two follow-up studies [6,8] with

similar distribution od Dukes’ stage was made. The

protocols, including physical examination, endoscopy,

imaging and tumour marker estimation were similar. In

the first, 639 patients were followed for 10 years with a

recurrence rate of 34%. About half of recurrences were

diagnosed within the first year after surgery and 90% of

Table 4 Results of FDG-PET in the

diagnosis for recurrence after colorectal

cancer.

Patient

No.

PET finding

(Abnormal FDG uptake)

Dukes

stage

True

positive

False

positive Validation

3 Liver C2 1 CT, FNAB†

10 Pelvis B2 1 Surgery

14 Liver C1 1 CT, FNAB,

histology, surgery*

23 Intra-abdominal C1 1 Surgery

29 Liver and Lung C2 2 CT, FNAB*

CT, computed tomography; FNAB, fine needle aspiration biopsy; CT performed

*6 months and †1 years after FDG-PET.

Figure 1 Multiple liver metastases confirmed by FNAB and

CT-Scan.

Figure 2 False positive pelvic FDG-PET scan.

Figure 3 Metastasis in the great homentum.

Figure 4 Lung metastasis confirmed by FNAB and CT-scan.

FGD-PET in recurrent colorectal cancer F. Selvaggi et al.

498 � 2003 Blackwell Publishing Ltd. Colorectal Disease, 5, 496–500

patients with recurrence died within 2 years. Only 30

(14%) of 218 patients with a recurrence underwent any

further curative procedure with salvage after re-operation

of 23% [8].

In the second, 363 patients were followed three

monthly for the first 2 years and 6 monthly for the next

3 years [6]. The recurrence rate of 33% was almost

identical. Fifty-eight percent of recurrences were detected

at regular follow-up examination and 72% of patients

with recurrence died during the follow-up period. As

with the first study only 13% of patients were re-operated

for cure.

It was concluded that an intensified follow-up leads to

earlier diagnosis of recurrence than standard follow-up

but salvage and survival were not improved by this

approach. Even with the development of new imaging

techniques there has been only a slight improvement in

the early diagnosis of recurrence especially in the pelvis.

Both MRI and CT scan appear to have a low specificity

with difficulty, especially for CT, in differentiating scar

fibrosis from recurrence [8–12].

In an early study of FDG-PET, 15 patients with an

abnormal mass after surgery for rectal cancer shown on

CT, underwent pelvic FDG-PET [3]. Of these, 11

(73%) patients had a positive scan confirmed by surgery,

biopsy and sequential CT scanning. A lesion was

considered to be scar fibrosis if it showed no change

in size and the patients showed no clinical symptoms for

one-year [3].

More recently 59 consecutive patients with suspected

recurrence underwent FDG-PET at a mean of 16 months

(range 0–119 months) after initial surgery. FDG-PET

showed an overall sensitivity of 100% and a specificity of

67% with positive and negative predictive values of 92%

and 100%, respectively. There were no false negatives but

5 patients were initially classified as false positive owing to

a negative CT scan, MRI or laparotomy, but were

subsequently found to be true-positive several months

later.

CEA and PET were compared in 28 patients who

underwent re-laparotomy after increased CEA levels

without evidence of disease or resectable disease on CT

scan, bone scan, colonoscopy, and MRI. All patients

had a FDG-PET and CEA-scan. Two surgeons, unaware

of the results of FDG-PET and CEA-scans, conducted

the exploration. Twenty-six patients had recurrent

disease; of these, 10 had unresectable disease with a

prediction of unresectability on FDG-PET in 9 (90%)

cases. FDG-PET scan predicted resectability in 81%

compared with 13% on the CEA-scan [13]. In a study

of 23 patients with a total of 25 recurrent lesions, FDG-

PET detected 15 of 16 histologically proven local

recurrences [5].

A recent meta-analysis showed that at equivalent

specificity, FDG-PET is the most sensitive noninvasive

imaging modality for the diagnosis of hepatic metastases

from colorectal and other gastro-intestinal tumours

compared with MRI, CT scan and Ultrasound [14]. If

survival after multimodal treatments for recurrence is to

be improved, early detection of recurrence should be

aimed for [15–18].

Assessment of the abilityof PET scanning to determine

operability has been made. Of 79 patients with known or

suspected recurrence the predictive accuracy of resecta-

bility by conventional imaging and FDG-PET was

compared. Recurrence was demonstrated in 68 patients

and FDG-PET was more accurate in predicting resecta-

bility (82% vs. 68%, P ¼ 0.02) [19].

In another study, 102 patients were evaluated by

FDG-PET for suspected or confirmed recurrence and

without evidence of unresectable disease on conventional

imaging modalities. PET-FDG lead to a change in the

management of 60 (59%) patients. Significantly,

re-operation was avoided in 26 patients showing a real

benefit of FDG-PET in avoiding inappropriate treatment

[20].

These results are confirmed by others [21]. Of 42

patients with suspected recurrence FDG-PET was more

accurate than CT for staging local recurrence (sensitivity

100%, specificity 86% vs. 75% and 100%, respectively) and

liver metastases (sensitivity 100% vs. 45%; specificity 100%

for both). Overall, FDG-PET upstaged 8 (27%) of 30

patients and altered patient management in 16 (38%)

patients [21]. These observations are in agreement with

the results of our study in which 2 of 3 patients with a

FDG-PET true-positive result who underwent surgery

had a macroscopically curative resection that was not

shown on CT.

Thus, FDG-PET has a clear role in the diagnosis of

recurrence with higher accuracy, specificity and sensitivity

compared with conventional methods. Moreover, FDG-

PET can alter management modifying surgical and

multimodal approaches and avoid unnecessary treatment.

Conclusions

The evidence from the literature and the results of the

present study suggesta new role for FDG-PET in the

routine follow-up of patients after curative surgery.

Recurrence usually becomes evident within one year of

surgery when CT and MRI may give a false negative

results. For this reason FDG-PET should be considered

in the routine postoperative assessment at 12 months. In

patients with clinical suspicion, abnormal abdominal

and ⁄ or pelvic masses or with increased tumour markers,

FDG-PET assessment is the first choice of imaging.

F. Selvaggi et al. FGD-PET in recurrent colorectal cancer

� 2003 Blackwell Publishing Ltd. Colorectal Disease, 5, 496–500 499

References

1 Galandiuk S, Wieand HS, Moertel CG et al. Patterns of

recurrence after curative resection of carcinoma of the colon

and rectum. Surg Gynecol Ostet 1992; 174: 27–32.

2 Thoeni RF. Colorectal Cancer. Radiological staging. Radiol

Clin North Am 1997; 35: 457–65.

3 Ito K, Kato T, Tadokoro M et al. Recurrent rectal cancer and

scar: differentiation with PET and MR imaging. Radiology

1992; 182: 549–52.

4 Moertel CG, Fleming TR, Macdonald JS, Haller DG, Laurie

JA, Tangen L. An evaluation of the carcinoembryonic

antigen (CEA) test for monitoring patients with resected

colon cancer. J Am Med Assoc 1993; 270: 943–7.

5 Takeuchi O, Saito N, Koda K, Sarashina H, Nakajiama N.

Clinical assessment of positron emission tomography for the

diagnosis of local recurrence in colorectal cancer. Br J Surg

1999; 86: 932–7.

6 Tornqvist A, Ekelund G, Leandoer L. The value of intensive

follow-up after curative resection for colorectal carcinoma.

Br J Surg 1982; 725–8.

7 Maurea S, Mainolfi C, Bazzicalupo L et al. Imaging of

adrenal tumors using F-18 FDG PET. comparison of benign

and malignant lesions. Am J Roentgenol 1999; 173: 25–9.

8 Berge T, Ekelund G, Mellner C et al. Carcinoma of the colon

and rectum in a defined population. Acta Chir Scand

Supplement 1973; 438: 1–84.

9 Moss AA, Thoeni RF, Scnider O et al. Detection and staging

for recurrent rectal carcinomas. J Comput Assist Tomogr

1981; 5: 870–4.

10 Mendez RJ, Rodriguez R, Kovacevich T, Martinez S,

Moreno G, Cerdan J. CT in local recurrence of rectal

carcinoma. J Comput Assist Tomogr 1993; 17: 741–4.

11 Rafto SE, Amendola MA, Gefter WB. MR imaging of

recurrent colorectal carcinoma versus fibrosis. J Comput

Assist Tomogr 1988; 12: 521–3.

12 Ruhlmann J, Schomburg A, Bender H et al. Fluorodeoxy-

glucose whole-body positron emission tomography in colo-

rectal cancer patients studied in routine daily practice. Dis

Colon Rectum 1997; 40: 1195–204.

13 Libutti SK, Alexander HR Jr, Choyke P et al. A

prospective study of 2-[18F] fluoro-2-deoxy-D-glucose ⁄positron emission tomography scan, 99mTc-labeled arci-

tumomab (CEA-scan), and blind second-look laparotomy

for detecting colon cancer recurrence in patients with

increasing carcinoembryonic antigen levels. Ann Surg

Oncol 2001; 8: 779–86.

14 Kinkel K, Lu Y, Both M, Warren RS, Thoeni RF. Detection

of Hepatic Metastases from Cancers of the Gastrointestinal

Tract by Using Noninvasive Imaging Methods (US, CT, MR

Imaging, PET): a Meta-Analysis. Radiology 2002; 224:

748–56.

15 Shirouzu K, Isomoto H, Kakegawa T. Total pelvic exenter-

ation for locally advanced colorectal carcinoma. Br J Surg

1996; 83: 32–5.

16 Bussieres E, Gilly FN, Rouanet P et al. Recurrences of rectal

cancers: results of a multimodal approach with intraoperative

radiation therapy. French Group of IORT. Intraoperative

Radiation Therapy. Int J Radiat Oncol Biol Phys 1996; 34:

49–56.

17 Shoup M, Guillem JG, Alektiar KM et al. Predictors of

survival in recurrent rectal cancer after resection and intra-

operative radiotherapy. Dis Colon Rectum 2002; 45: 585–

92.

18 Salo JC, Paty PB, Guillem J, Minsky BD, Harrison LB,

Cohen AM. Surgical salvage of recurrent rectal carcinoma

after curative resection: a 10-year experience. Ann Surg

Oncol 1999; 6: 171–7.

19 Lonneux M, Reffad AM, Detry R, Kartheuser A, Gigot JF,

Pauwels S. FDG-PET improves the staging and selection of

patients with recurrent colorectal cancer. Eur J Nucl Med Mol

Imaging 2002; 29: 915–21.

20 Kalff V, Hicks RJ, Ware RE, Hogg A, Binns D, McKenzie

AF. The clinical impact of (18) F-FDG PET in patients with

suspected or confirmed recurrence of colorectal cancer: a

prospective study. J Nucl Med 2002; 43: 492–9.

21 Arulampalam T, Costa D, Visvikis D, Boulos P, Taylor I, Ell

P. The impact of FDG-PET on the management algorithm

for recurrent colorectal cancer. Eur J Nucl Med 2001; 28:

1758–65.

FGD-PET in recurrent colorectal cancer F. Selvaggi et al.

500 � 2003 Blackwell Publishing Ltd. Colorectal Disease, 5, 496–500