Surgery is the main treatment with curative potential for recurrent and metastatic (mainly liver) colorectal cancer. The presence of the disease at a site distant to the planned surgery affects the type and timing of treatments. Together, this wide variation in disease presentations and extents of treatment underpins the rationale for accurate pre-operative staging. Although colonoscopy is the most common method for detecting and diagnosing colorectal cancers, it does not produce accurate preoperative information regarding tumor invasion and lymph node involvement. For this purpose, MDCT and MRI are used as the standard modality for preoperative staging of colorectal cancers. Moreover, neoadjuvant therapies are performed for stage II and stage III rectal cancer. It has been shown that neoadjuvant therapy decreases local recurrence and increases survival . Therefore, correct preoperative staging has a critical role in determining whether patients should undergo neoadjuvant therapy. Similarly, preoperative neoadjuvant strategies for colonic cancers are dependent on the staging accuracy of CT . Unfortunately, CT cannot provide complete correct staging of colorectal cancer, even though an improvement in the resolution has recently been achieved with MDCT.
Use of MDCT on determination of T and N staging has variable results [5, 13–16]. Although it has been reported up to 86% of accuracy for T staging, this rate has been decreased to 59% for N staging. Lack attenuation differences between tumor and normal visceral softy tissue, and inadequate distension of the bowel are thought to be responsible for low sensitivity and specificity of MDCT for T staging . MDCT can be considered to be more efficacious for N staging than T staging. However, efficacy of MDCT on determination of lymph node status of CRC has not been showed by many studies. Due to heterogeneity on the design of the studies and technical differences, it has been impossible to get generally accepted results.
Likewise, MRI offers an 81% overall accuracy compared with histological findings for T and N staging . Due to these imperfect results of CI studies, authors have investigated new preoperative imaging modalities for colorectal cancer; some authors suggest the use of PET/CT as an alternative option [15, 18, 19]. It has been shown that the accuracy of PET/CT can be as high as 94.3% for T staging. However, sensitivity and specificity for N staging still remains to be low in comparison to T staging . Positive lymph nodes that are smaller than 1 cm may be the major source for being missed by PET/CT . In this study, it has been shown that PET/CT has the accuracy rate of 90.47% for T staging. Although it can be regarded as an acceptable rate for T staging, lack of both pathological confirmation and comparison with CI studies attenuate the reliability of our results.
In addition to provide accurate staging, the ability of PET/CT to detect metastatic disease is thought to be a critical point for its potential therapeutic impact. However, studies have reported inconsistent findings about the effect of PET/CT on clinical practice and surgical management. Some studies found no effect and others reported decreased morbidity from improved surgical techniques arising from increased precision in tissue identification. In some studies, PET/CT was compared directly with CT alone, which is one example of a falsely enhanced apparent therapeutic impact for PET/CT. PET/CT has commonly been performed for the detection of recurrence or for routine follow-up in patients with colorectal cancer. The sensitivity and specificity of PET/CT in patients with recurrent colorectal cancer was found to be 97% and 76%, respectively . However, PET imaging provides insufficient anatomical information; this lack of information was improved with the integration of CT into PET imaging. Cohade et al. showed that the accuracy of PET alone and PET/CT in preoperative staging of colorectal cancer was 78% and 89%, respectively . MRI also provides additional accuracy to liver contrast-enhanced CT in the assessment of a patient’s suitability for hepatic resection.
There are limited studies in the literature that have investigated PET/CT in the preoperative staging of colorectal cancer. By using PET/CT as a preoperative imaging modality, it was reported that the stages of the tumors were changed in 27% to 39% of the patients either down- or over-stages [21–24]. Our data showed that the preoperative stage changed in 21.9% of the cases according to the PET/CT results; this result is comparable to findings in the literature. However, a modification of clinical management including only the surgical treatment modality was found in only 3.2% of the cases. In eight and six cases, there were down and over staging of the primary tumor which have no effect on the choice of surgical treatment, respectively. Any modifications with regard to postoperative adjuvant treatment caused by preoperative PET/CT were beyond the primary aim of this study, which was to evaluate the effect of PET/CT on surgical treatment. It has been reported that use of PET/CT in staging of rectal cancer resulted in discordant and incidental findings in an almost half of the cases. Although PET/CT brought on stage migration in 30% of the cases, either down or over staging, potential management changes occurred only in 25% and there was no need to change the surgical management . This discordance may be explained by a high false positivity rate of PET/CT for detecting distant metastasis. PET/CT detected 6 distant metastases, which could not be shown by CI studies. Four of the 6 distant metastases were verified as false positive by biopsies (Table 5). The high rate of false positive results which was caused by the high accumulation of FDG in mediastinal lymph nodes and hilar region might be related to the high incidence of chronic infectious and inflammatory diseases of the chest in our country. The present study suggests that some additional evidence offered by PET/CT was not always beneficial and caused preoperative diagnostic dilemmas, which caused further invasive examinations, additional costs and a delay of the disease management. Selecting patients with locally advanced tumors in whom distant metastases are more expected and patients with suspected metastasis detected with other imaging modalities for PET/CT scanning may be more beneficial and practical for clinical use.
Local staging of colorectal cancer mostly depends on CI studies. There are various studies regarding CT and MRI that reported a high success rate. The specificity and sensitivity of CT and MRI for the detection of adjacent organ invasion were reported as comparable . PET/CT is inappropriate to determine the exact depth of invasion of the primary tumor due to its limited resolution. However, PET/CT may be appropriate in selected cases to estimate penetration and local invasion. MDCT provides more accurate anatomical and structural information than PET. Therefore, T staging of colorectal cancer by PET/CT is almost completely reliant on CT. As expected, the present study demonstrated a close correlation of PET/CT with CI studies in T staging (Tables 1 and 3).
The main problem of staging of colorectal cancer is the prediction of lymph node involvement. The sensitivity of CT for the detection of lymph node involvement has been reported to be between 29% and 90% for CRC [13, 25]. PET/CT showed low sensitivity (52%) and relatively high specificity (85%) for detecting lymph node involvement in the present study. The overall accuracy of PET/CT (63%) was below the expected value. Several previous studies have reported a comparable rate of lymph node involvement detection by PET/CT, reporting low sensitivity (29–37%) and high specificity (83–96%) [25, 27, 28].
A full assessment of the colon is mandatory to localize the tumor, to evaluate locoregional spreading, and to depict synchronous colonic lesions. For that purpose, MDCT is the most favored imaging technique. However, CT might have poor performances for determining local tumor extension in the absence of colon distension . CT colonography is another exam which is primarily used to evaluate the colon in cases of incomplete colonoscopy and as an alternative means of screening for colorectal carcinoma [30, 31]. Although both contrast enhancement by using intravenous agents to define the boundaries of structures, and colonography to identify primary tumor with its local extent increases the accuracy of such modalities, the choice of PET/CT without intravenous contrast medium or colonography has been shown to be effective, especially in T staging of CRC . Moreover, contrast-enhancement causes more accurate N staging of rectal cancer compared with non-contrast-enhancement during PET/CT examination. PET/CT colonography is also used in preoperative diagnosis of the tumors proximal to obstructive colorectal cancers, which were defined as cancers that cannot be traversed colonoscopically . Use of this technique has been reported to have an overall accuracy of 80% and 60% for the evaluation of tumor depth and lymph nodes, respectively. Use of water enema or air-contrast enema during CT colonography may also result in better evaluation of the local spread for T staging for CRC [31, 33]. However, FDG as a radiotracer may play the role of “metabolic contrast agent”. By that way, it can be helpful to increase the contrast resolution of the structures, to characterize the perilesional tissues and to compensate for the absence of luminal distension on the unenhanced CT images . Therefore, identification of the primary tumor with its local extent by using FDG PET/CT, without administration of intravenous contrast medium or colonography can be possible as supported by our results.
The high false-negative rate of PET/CT may be attributable to the limited resolution and proximity of the involved lymph nodes to the locally advanced primary tumor or the urinary bladder. In addition, while a lymph node with a micro-metastasis and a diameter >5 mm can be considered to be involved by CT assessment, the same lymph node can be considered as non-metastatic by PET/CT because no FDG uptake is detected. These results imply that preoperative PET/CT is of limited value for detecting metastasis to regional lymph nodes.
There were four suspected pulmonary and one supraclavicular metastasis in our patients. Although chest CT before PET/CT could help to differentiate the malignant potential of such lesions, use of chest CT in staging of CRC remains controversial [17, 34]. It was shown that chest CT altered the initial TNM stage in less than 1% of CRC patients. In addition, indeterminate lung nodules were found to be positive in almost one quarter of the patients . In the light of these findings, chest CT was not used as the primary staging method in this study.
The main limitation of our study was the wide variety of pathological groups and subgroups of the primary tumors that may have influenced the sensitivity and accuracy of PET/CT for TNM staging of colorectal cancer, as FDG uptake may differ among tumor types. Although the contribution of PET/CT to the detection of recurrent and metastatic colorectal cancer has been reported in many current studies, its value in staging the primary disease has not been well-defined and usually is not recommended as a first-line diagnostic tool in clinical practice. Although most of the tumors presented in this study had T staging of T2 and T3, and N staging of N0 and N1, it could be difficult to generalize the results to all subgroups of colorectal cancers. In addition, presence of both colonic and rectal cancers, and application of the neo-adjuvant treatment to Stage II and III rectal cancers might be the other confounding variables to affect the reliability of our results. Therefore, more studies that include special subgroups of colon and rectal cancers are necessary to determine the role of PET/CT in primary staging.
PET/CT seems to be a useful tool in the evaluation of colorectal cancer by allowing to metabolically characterizing undetermined lesions suspected for recurrence of disease, to perform a complete pre-surgical staging and to identify occult metastatic disease. However since it is an expensive modality and the impact to the management of disease may be low as in our study; its use in routine preoperative examination is controversial.
Another aspect to be considered for the routine use of PET/CT as a first-line diagnostic modality is contrast-enhanced PET/CT scanning, which may replace the routine contrast-enhanced CT imaging; this method will allow for whole body detection of distant metastases and show the primary tumor and loco-regional lymph nodes more accurately. Studies with contrast-enhanced PET/CT scans are needed in the future.