Indian Journal of Cancer, Vol. 40, No. 3, (July - September 2003) , pp. 87-100
Positron Emission Tomography imaging in Evaluation of Cancer Patients
Kumar R, Bhargava P, Bozkurt MF, Zhuang H, Potenta S, Alavi A
Division of Nuclear Medicine, Department of Radiology,
Hospital of the University of Pennsylvania, 110 Donner Bldg., 3400 Spruce
Street, Philadelphia, PA, 19104, USA.
Code Number: cn03015
Positron emission tomography (PET) is a diagnostic imaging technique that has progressed rapidly from being a research technique in laboratories to a routine clinical imaging modality. The most widely used radiotracer in PET is Fluorine18-fluorodeoxyglucose (F18-FDG), which is an analogue of glucose. The FDG uptake in cells is directly proportional to glucose metabolism of cells. Since glucose metabolism is increased many fold in malignant tumors PET has a high sensitivity and a high negative predictive value. PET with FDG is now the standard of care in initial staging, monitoring the response to the therapy, and management of lung cancer, colonic cancer, lymphoma, melanoma, esophageal cancer, head and neck cancer and breast cancer. Other indications of PET like bone tumor, ovarian cancer and cancer of unknown primary (CUP) has also been discussed in brief. The aim of this review article is to review the clinical applications of PET in various malignancies and only limited number of important studies will be discussed for this effort.
Key Words: Positron emission tomography, Fluorodeoxyglucose, Lymphoma, colonic cancer, Lung nodule and lung cancer, Breast cancer, Head and neck cancer, Ovarian cancer, Melanoma, Gastric and esophageal cancer.
Positron emission tomography (PET) is an advanced diagnostic imaging technique that is used for diagnosis, staging and restaging after treatment or recurrence of various cancers. This technique exploits the increased metabolism of glucose in malignant viable cells. 2-[fluorine18-fluoro-2-deoxy-D-glucose (FDG) which is an analogue of glucose detects the difference in glucose metabolism. Like glucose, FDG is transported into tumor cells, by means of glucose transporter protein and subsequently it is phosphorylated by an enzyme hexokinase to FDG-6-phosphate.1,2 As FDG-6-phosphate is not a substrate for glucose-6-phosphate isomerase (the next step in glycolysis), it is biochemically trapped within the cell. This process of metabolic trapping in cell constitutes the basis for imaging of distribution of tracer with PET. Since there is many fold increase in glucose metabolism in malignant tumors as compared to normal, it is easy to detect this difference in metabolism using PET. Therefore PET has high sensitivity and high negative predictive value as compared to conventional morphological modalities like computed tomography (CT), Ultrasonography (USG), magnetic resonance imaging (MRI), which use size as the only or major criterion to distinguish between benign and malignant disease in lymph nodes.3,4 PET has well-established role in initial staging, monitoring the response to the therapy, and management of various cancers (Table 1). The list of reimbursable indications for FDG PET is growing, as more research is being done.
Solitary Pulmonary Nodules
The incidence of malignancy in solitary pulmonary nodules (SPN) ranges from 10% to 70% and there is still controversy regarding optimal management of solitary pulmonary nodules.5,6 CT is useful mainly in the correct localization of SPNs and it can also differentiate nearly one-third of SPNs as malignant or benign in etiology, based on image characteristics.7,8
FDG-PET has been successfully used as a non-invasive diagnostic test for SPNs in order to differentiate benign lesions from malignancy. In the visual interpretation of FDG PET images, intense uptake of FDG at approximately at 1-hour images after injection is considered `positive' (Figure 1). Recent studies have shown better results with delayed images taken several hours after injection (dual time point imaging), along with routine 1-hour imaging protocol.9-11 These studies concluded that malignant nodules show increasing FDG uptake by time, while pulmonary nodules of benign origin have a declining pattern of uptake with time. FDG uptake in a lesion can be quantitated by calculating its SUV (standardized uptake value). In general, SUV over 2.5 is considered as characteristic of malignancy12,13 whereas in some studies SUV over 2 has also been used as a threshold value.14
The importance of PET for the differential diagnosis of SPNs has been shown in many studies.15-18 A meta-analysis study in which 450 SPNs were included concluded that the mean sensitivity and specificity for detecting malignancy were 93.9% and 85.8% respectively.19 The studies that investigated the role of FDG-PET imaging in the differential diagnosis of SPNs reported a wide variation in their specificity. The cause of this wide variation may be attributed to the prevalence of granulomatous diseases like tuberculosis, histoplasmosis or coccidioidomycosis which are varied widely according to the geographical localization.
FDG-PET has very important role in the management of patients with SPNs, but use of FDG-PET for this purpose warrants a high negative predictive value, which is related to the pre-test probability of cancer and to the sensitivity and specificity of FDG-PET method. Gambhir et al reported that FDG-PET imaging is an appropriate method for detection of malignancy in SPNs only for patients with 50% or lower pre-test probability of cancer, as determined by factors like lesion size, age, smoking history.20 If the patient has a higher pre-test probability for cancer, histologic diagnosis is needed regardless of FDG-PET result, since the post-test probability of cancer after a negative PET study may be as high as 10%.20 It is recommended that patients with a negative PET study and a high pre test probability should be followed up by clinical assessment and imaging studies, since a little risk of false negative result remains. For these patients a follow-up of 6-12 month intervals for monitoring the possible increase in the lesion size will be an indication for histologic diagnosis.19,20
FDG-PET has been approved differential diagnosis of solitary pulmonary nodules. Since FDG-PET has a high diagnostic accuracy, it has been incorporated into the routine diagnostic procedures for differentiation of benign nodules from malignant ones.
The diagnosis of lung cancer still remains to be a problematic issue. Morphological imaging methods such as CT of chest can detect a nodule or mass in the lungs but cannot differentiate benign from malignant lesions. Because of this, accurate diagnosis of lung cancer can only be established with invasive diagnostic procedures, that are associated with complications and that have a morbidity rate of 1-10%.21
In the literature the majority of FDG PET studies for lung cancer have been done in non-small cell lung cancer (NSLC) patients, in whom regional and distant involvement of disease will change staging and guide the therapeutic approach. In general, CT is the method of choice to define the extent of the primary tumor and to assess the tumour involvement of the pleural surfaces and the thoracic wall.22 For some patients, MRI can be used as an additional imaging modality, especially to evaluate the relationship of tumor to the heart and large vessels.23,24 The main limitation of CT/MRI is using a size criterion of 1 cm for the diagnosis of tumour involvement. For this reason, CT cannot report any lymph node with tumor involvement if it is smaller than 1 cm in diameter. In contrast, all of the lymph nodes larger than 1 cm in diameter will be defined as tumor involvement and will mislead the therapeutic management. Since FDG PET imaging depends on the criterion of increased metabolism, but not size, it can be used successfully in order to differentiate malignant nodes from the others (Figure 2). In the literature there are some studies in which the accuracy of CT and FDG PET have been compared to each other.25,26 Pieterman et al reported that the sensitivity and specificity of PET and CT for the evaluation of mediastinal nodal involvement were 91% and 86% for PET; and 75% and 66% for CT respectively.26 Dwamena et al in their meta-analysis of staging lung cancer by PET and CT concluded that FDG PET was significantly more accurate than CT and reported the sensitivity and specificity values of 79% and 91% for PET and 60% and 77% for CT respectively.27 Several authors reported that correlation of anatomic (CT, MRI) imaging with functional (PET) imaging or PET/CT fusion imaging may also be useful for the further improvement of determining the patients with nodal disease.28,29
PET also can detect distant metastases effectively.30-33 FDG PET can detect distant metastases that are unsuspected after conventional staging in 10-14% of lung cancer patients.26 FDG PET is an accurate imaging modality in order to rule out metastasis in cases of false positive or indeterminate anatomical imaging findings. FDG PET has also a particularly important role for detecting adrenal metastases in lung cancer patients. Up to 20% of lung cancer patients are found to have an adrenal mass by CT, without necessarily confirming metastasis. It has been reported that FDG PET can eliminate the need for biopsy of enlarged adrenal glands in lung cancer patients.34,35
Monitoring Treatment Response
Anatomical imaging methods use the criterion of tumor shrinkage in order to define a good therapy response but this criteria may not be the best indicator of response to therapy, since fibrotic tissue secondary to therapy may give the same image pattern as a non or partially responding mass. FDG PET imaging, which uses metabolic criterion, is a more accurate method to differentiate tumor from scar tissue. Bury et al reported that FDG PET was more sensitive and equally specific compared to other imaging modalities for detection of residual disease or recurrence after surgery or radiotherapy.35 In the literature there are also some other studies reporting the higher accuracy of FDG PET for differentiating treatment changes from recurrent disease compared to CT.35,36 Also a cutoff SUV of 2.5 may be accepted as accurate in differentiating tumor from benign post therapeutic changes.37
Normalization of FDG uptake following the therapy has been reported to be a good prognostic sign.38.39 Hebert et al showed that all of the patients in their study with normal FDG uptake after therapy were alive at 2 years after treatment, while 50% of patients with post therapeutic residual FDG uptake had died within the same 2 year period.39 In the literature some studies reported an independent prognostic value of FDG uptake in lung cancer, so that patients with SUV of more than 7 or 10 had shorter survival compared to the ones with SUV below these thresholds.37,40
FDG-PET imaging has been approved for the diagnosis, initial staging and restaging of only non-small cell lung cancer. Although there are some studies in the literature that provides convincing evidence for the use of FDG-PET for small cell lung cancer, further studies are still warranted.
Colorectal carcinoma is the third most common malignancy in men, and the second in women and is the fourth leading cause of all cancer deaths.41,42 If the disease can be diagnosed in the early stages, the surgical treatment is almost curative with minimal morbidity and mortality rates. Only 70-80% of patients can undergo curative resection at presentation and their overall 5-year survival is less than 60%.43
Although the sensitivity of FDG PET for the preoperative diagnosis of colorectal carcinoma is high, it has no important role in practice, since surgical diagnosis and staging will be needed for all patients. Abdel-Nabi et al reported that FDG PET imaging was successful for the identification of all primary lesions in 48 patients.44 However both FDG PET and CT missed approximately 70% of nodal metastases. But FDG PET was found to be superior to CT for the identification of liver metastases with a sensitivity and specificity of 88% and 100% respectively versus 38% and 97% respectively for CT. The common false positive findings of FDG PET were fistulas, abscesses, diverticulitis and less frequently adenomas.
Monitoring Treatment Response
Haberkorn et al reported the efficacy of FDG PET to monitor the response to radiation therapy.45 In this study, statistically significant reduction in FDG uptake of tumors following radiation therapy was seen only in 50% of patients, and reduction of FDG uptake correlated with satisfactory palliative results. FDG PET was also found to be more sensitive in detection of recurrent disease compared to serum CEA measurement. Vitola et al studied the effects of regional chemoembolization therapy using FDG uptake as a criterion and found that decreased FDG uptake correlated with response while presence of residual uptake was used to guide further regional therapy.46
In their study Findlay et al assessed the metabolic characteristics of colorectal cancer liver metastases in 18 patients using both FDG PET and CT before and during the first month of 5-Fluorouracil chemotherapy, regarding to the changes in size and FDG uptake of lesions.47 They performed FDG-PET studies before treatment, after 1-2 week and 4-5 week of treatment. They concluded that FDG PET was accurate for differentiation of responders from non-responders, both on lesion-by-lesion or patient-by-patient analysis. Tumor response was associated with lower tumor-to-liver uptake ratios at 1-2 week and 4-5 week FDG PET studies, as well as lower SUVs after 4-5 weeks of therapy.
Colorectal cancer will recur in approximately 30% of patients within 2 years, and in 25% of these patients the recurrence of disease will be isolated to one site, mostly liver and surgery may serve as a potentially curable therapeutic option.48 The presence of extrahepatic metastases is considered to be a contraindication to lesion excision, and also the number and the size of hepatic metastases adversely affect prognosis. FDG PET has an important role to detect extrahepatic metastases and plays a vital role in selection of patients who may benefit from partial hepatectomy.
In general, serial serum CEA levels are used for recurrence monitoring, but when a high serum level of CEA is encountered, imaging will be necessary to localize the site of possible recurrence. CT is usually incapable of differentiating post surgical changes from recurrence and also it commonly misses extrahepatic abdominal metastases and fails to detect hepatic metastases in 7% patients as well.49 Functional imaging with FDG PET can be successfully used to identify the metabolic characteristics of the lesions, which are equivocal or even not seen by CT, and FDG PET can define the recurrence as a complementary test (Figure 3). Valk et al in their study compared the sensitivity and specificity of FDG PET and CT side-by-side.50 According to their study, FDG PET was found to be more sensitive for all sites except the lung, where both imaging techniques had equivalent sensitivities. They also reported that one third of FDG positive lesions in the abdomen, pelvis and retroperitoneum were negative on CT. Other studies showed that whole-body FDG PET imaging is especially useful for defining distant metastases, such as abdominal nodal involvement and lung metastases.51,52 Also PET is accurately capable of differentiating post surgical changes from recurrence, in contrast to CT, the accuracy of PET in this regard ranges from 90-100% where accuracy values for CT are reported to be 48-65%.54
FDG-PET has been approved for diagnosis, initial staging and restaging of colorectal cancer. However FDG-PET is not practically preferred for the preoperative diagnosis of colorectal cancer. Currently the main indication of FDG-PET for colorectal cancer is the assessment of regional and distant metastasis and mainly the identification of recurrence in patients with rising CEA levels when conventional imaging modalities fail.
Breast cancer is the most common malignancy in women in most of countries. The use of F18-FDG PET in breast carcinoma has been evaluated in the initial staging, monitoring the response to chemotherapy and determination of metastatic disease.
In the last several years, there have been intense efforts to apply various imaging modalities to improve the detection, diagnosis and local staging of breast cancer. Wahl et al and Bruce et al detected all primary tumours using PET in 10 and 15 patients respectively.55,56 FDG PET plays an important role in differentiation between benign and malignant tumours.57-59 The sensitivity of PET for breast cancer varies between 68% and 100%.55-59 A study by Schirrmeister et al showed that whole body FDG PET is as accurate as panel of imaging modalities currently employed and significantly more accurate in detecting multifocal disease, lymph node involvement and distant metastasis (Figure 4).59 Same group evaluated relation of FDG uptake and biological and clinical prognostic parameters and found that FDG uptake was significantly higher in ductal than in lobular cancer and FDG uptake correlates with proliferative activity assessed by Ki-67 immunostaining.60
Metastasis to axillary lymph nodes is one of the most important prognostic factors in breast cancer patients. There is no effective non-invasive modality to know the involvement of axillary status of lymph nodes except axillary dissection and sentinel lymph node sampling. Cripa F et al showed overall diagnostic accuracy of 86%. The accuracy was very high (95%) in N1a patients.61 The sensitivity of FDG PET in detection of axillary lymph node metastasis varies from 79-100%.61-63 FGD PET can miss micrometastases in lymph nodes as there are fewer cells, which may or may not have increased glucose metabolism to be detected. Axillary lymph nodes dissection or sentinel lymph node samplings are the best method to know the axillary lymph node involvement. 64
Monitoring Treatment Response
Multi-modality therapy beginning with neoadjuvant chemotherapy (NACT) is the standard of care for the patients with locally advanced breast cancer (LABC). NACT is associated with a good response rate in more than 70% of the patients including a complete pathologic remission rate of about 10% to 15%.65,66 Following successful induction chemotherapy, a variety of options for local treatment have been investigated. These include surgery alone, surgery and radiotherapy, and radiotherapy alone.67,68 However, unless a reliable method is available to assess the residual tumor after neoadjuvant chemotherapy, decisions on the need and the extent of subsequent surgical management would be difficult.
FDG PET has been reported to detect metabolic changes in breast cancer as early as 8 days after initiation of chemotherapy.69 In the literature there are several studies, which differentiated responder from non-responder after first course of therapy using FDG PET imaging.70-72 Responders were correctly identified with sensitivity of 90% and 100% and specificity of 74% and 85% by Schelling et al and Smith et al respectively after first course of chemotherapy.71,73 Bassa et al found that FDG PET was useful in evaluating response to chemotherapy for primary LABC.72 They showed a good correlation of persistent FDG uptake after completion chemotherapy and poor prognosis. In a recent study, the factors influencing the response to chemotherapy in LABC patients were investigated using FDG and O15-water.74 The authors hypothesized that low tumor perfusion and high FDG metabolism correlated with the poor response to chemotherapy.
Vranjesvic et al predicted the response to chemotherapy in 61 patients using FDG PET and conventional imaging (CT/MRI/USG).75 PET was more accurate than combined conventional imaging modalities with positive and negative predictive values of 93% and 84% respectively for FDG PET versus 85% and 59% respectively for conventional imaging modalities. The accuracy was 90% and 75% for FDG PET and conventional imaging modalities respectively.
Locoregional and distant metastasis occurs in up to 35% of patients within 10 year of initial surgery of breast cancer.76 Early detection of recurrent or metastasis may be beneficial in instituting local or systemic treatment to prevent tumor related symptoms.77,78 CT and other morphologically oriented images reflect mainly morphological changes and nodular size is the only or major critical criterion of malignancy. Whole body FDG PET has emerged as a sensitive and noninvasive technique in management of breast cancer. 79
Gallowitsch et al in a retrospective analysis in 62 patients, showed sensitivity, specificity, PPV, NPV and accuracy of 97%, 82%, 87%, 96% and 90% for FDG PET and 84%, 60%, 73%, 75% and 74% for conventional imaging.80 On lesion basis, significantly more lymph node84 vs 23 and fewer bone metastases61 vs 97 were detected by using FDG compared with conventional imaging. Kamel et al analysed the role of FDG PET 60 patients for detecting relapse.81 The overall sensitivity, specificity and accuracy were 89%, 84% and 87% for locoregional and 100%, 97% and 98% for distant metastasis. The author also concluded that FDG PET was more sensitive than the serum tumor marker CA 15-3 in detecting breast cancer relapse. Eubank et al compared FDG PET and CT in 73 recurrent/metastatic breast cancer patients for evaluation of mediastinal and internal mammary lymph nodes metastases. In 33 patients amenable to follow-up CT or biopsy, FDG PET revealed a superior detection rate of 85% compared to CT 54%, when prospectively interpreted.82,83
FDG PET has shown a variable degree of sensitivity and specificity in detecting bone metastasis. Moon et al reported a sensitivity of 93% and specificity of 79% with FDG PET in detecting recurrent or metastatic disease. FDG PET had many false negative results, particularly sclerotic lesions. Gallowwitsch et al and Kao et al had better specificity but lower sensitivity for detecting bone metastasis.80,84 Non-detectable lesions on FDG PET were sclerotic or mixed sclerotic/osteolytic radiologically. PET has also limited role in detecting bone metastasis in skull due to high uptake in brain. The combination of FDG PET scan and bone scan can be done for better sensitivity and specificity for bone metastases in breast cancer patients.
Currently, FDG-PET that have been approved for initial staging in patients with distant metastases, restaging in patients with locoregional and distant metastases and monitoring response to therapy in patients with locally advanced and metastatic disease. However, evaluation of regional lymph nodes is still not covered, since sentinel lymph node detection is the method of choice for this purpose.
Presently, most of the management decisions in lymphoma patients are based on CT scan for initial staging and follow-up, measurements of specific blood indices such as the ESR, absolute lymphocyte counts and serum LDH. However CT can only define anatomy and is insufficient for depiction of small volume of tumors in normal sized structures as size is the main diagnostic criterion of malignancy. It has also limited role in evaluating response to treatment, as most of the lesions does not shrink, especially fibrotic lymph nodes. Metabolic imaging with FDG PET has emerged as a powerful imaging modality for diagnosis, staging, and treatment monitoring of lymphoma patients.
Diagnosis and staging mainly depend on history, clinical examination, laboratory data, imaging with CT and/or MRI and Gallium-67. The accuracy of FDG PET is better than CT/MRI and Gallium scintigraphy.85,86 Moog et al studied 60 consecutive patients with HD and NHL, FDG PET found 25 additional lesions, while CT found 6 additional lesions of which three were false positive.87 Stumpe at al had similar experience when accuracy of FDG PET was compared with CT.88 There was no significant difference in sensitivity but the specificity was 96% for HD and 100% for NHL using PET, while the corresponding values were 41% and 67% for CT. Cremerius et al and Zinzani et al both showed a sensitivity of 100% and specificity of 17% for CT while sensitivity were 100% for both and specificity were 92% and 96% for PET.89,90 Recently, Sasaki et al showed a specificity of 99% for CT and PET but sensitivity was 65% and 92% for CT and PET.85 These studies show that there is large variation in sensitivity and specificity of CT, whereas these figures are not that variable with FDG PET (Figure 5). Another major advantage of FDG PET is that, it is a whole body imaging method.
A number of publications have posed the clinically relevant question of how FDG-PET impact on staging on patient management. PET modified therapy in 25% of all lymphoma patients. Apparently, 23% of patients were assigned a different stage by FDG PET imaging compared with conventional imaging.87,91.92
Monitoring Treatment Response
Several studies have shown the effectiveness of FDG PET in post treatment of lymphoma and found that PET has high predictive value for the differentiation between active tumor and fibrosis (Figure 6).93 Jerusalem et al compared the prognostic role of FDG PET and CT after first line treatment in 54 NHL/HD patients.94 FDG PET had higher diagnostic and prognostic value than CT (positive predictive value 100% vs 42%). The 1-year PFS survival was 86% in PET -Ve patients as compared to 0% in PET +Ve patients. Mikhaeel et al compared FDG PET with CT as prognostic indicator in treatment of aggressive NHL. The relapse was 17 % for PET -Ve and 100% for PET +Ve compared with 25% and 41% for negative and positive CT respectively.95 The PFS was 83% for PET -Ve as compare to 0% for PET +Ve group. Recently, Juweid et al also showed PPV and NPV for FDG PET for 1-year PFS were 92% and 88% respectively, compared with 47% and 85% for CT in patients with NHL.96
Spaepen et al evaluated 60 patients of HD who had FDG PET at the end of first line treatment with or with out residual mass. The 2-year DFS was 4% for PET +Ve and 85% for PET -Ve group.97 Mikhaeel et al confirmed these findings in 65 HD patients, with a 1-year PFS of 0% vs 93%.98
FDG PET also has important role in prognostication patients even after first cycle of chemotherapy. Kostakoglu et al recently compared FDG scan done after first cycle of chemotherapy and after completion of chemotherapy.99 PET had greater sensitivity and PPV after the first cycle (82% vs 45% and 90% vs 83%).
Approximately two third of patients with HD present with mass lesion in location of a previous tumor manifestation, but, only about 20% of patients ultimately relapse.100,101 Similarly, in patients with NHL 50% present with mass lesion and only 25% relapse in high grade NHL.101-102 A single CT or other morphological imaging cannot differentiate between fibrosis and recurrence. Since FDG uptake depends upon the metabolic state of the lesion in question and not on the size of the lesion, FDG PET provides solution to this diagnostic dilemma. Though Gallium scintigraphy has proved useful in patients with recurrent disease, it is less sensitive in intraabdominal and low-grade lymphoma.88,103
Cremerius et al reported sensitivity 88% and a specificity of 83% for detection of residual disease for FDG PET. The corresponding values for CT were 84% and 31% respectively.104 Recently, a study by Mikosch et al compared PET with CT/US in detecting recurrence and reported a sensitivity of 91%, specificity of 81%, PPV of 79%, PPV of 92% and accuracy of 85%. The similar values 88%, 35%, 48%, 81% and 56% respectively for CT/US.105
The present approved indications for FDG-PET are diagnosis, initial staging and restaging of Hodgkin`s disease and non-Hodgkin lymphoma. Based on the present literature data, FDG-PET has been found to have high diagnostic accuracy for staging and restaging lymphoma and it can provide better prognostic information compared to anatomical imaging modalities.
Head and Neck Cancer
The presence of lymph node spread of head and neck tumours is associated with substantially worse prognosis. Accurate pre therapy lymph node staging is essential for planning surgical or other management strategy. Conventional imaging like CT/MRI detects only fewer than 50% of involved nodes due to their dependence on size criteria and many times result in unnecessary neck surgery. Lymph node staging in these patients with FDG PET is more accurate than conventional imaging modalities.106-109 The sensitivity and specificity of FDG PET varies 71% to 91% and 88% to 100% in detecting metastatic lymph nodes.110 Adams et al and Kau et al reported sensitivity and specificity 90% and 94% for FDG PET, 80% and 85% for CT and 88% and 79% for MRI respectively.108-109 FDG PET can demonstrate up to 30% of previously undetected primary head and neck tumours (Figure 7).111
After radiation therapy and surgery, distortion of normal anatomic structures limits the ability of MRI and CT to demonstrate residual or recurrent disease. The FDG PET has comparable sensitivity of 88-100% vs 72-92% and better specificity of 75-100% vs 50-57% with CT and MRI in detecting recurrence at primary site.112-114 FDG PET is also more sensitive and specific in detecting residual and recurrent lymph node metastasis.113,114 The accuracy of FDG PET was 85% as compared to 42% for CT in differentiating recurrent laryngeal cancer from post radiation soft tissue changes.115 In identifying the response to treatment PET has very important role. A study by Lowe et al showed mean reduction of 82% FDG activity in patients who achieved complete remission compared to 34% reduction in FDG activity in those who had recurrent disease.116 The sensitivity and specificity of FDG PET for residual cancer 1-2 week after treatment were 90% and 83% respectively in patients with head and neck cancer.117 Recently, Brun et al predicted the therapy outcome using FDG PET in 47 head and neck cancers patients. They concluded that FDG PET in early phase of treatment of head and neck cancer is associated with tumour response, survival and locoregional control.118
At present FDG-PET has been approved for the diagnosis, initial staging and restaging of head and cancer, excluding the neoplasm's of the central nervous system and thyroid gland. Recent advances in imaging technology, such as the routine use of devices like PET/CT will probably increase the use of PET for head and neck cancer patients.
Many researchers evaluated the role of FDG PET for staging of melanoma (Figure 8). The accuracy of PET in detecting metastatic melanoma has ranged from 81-100%.119 But the sensitivity decreases when the tumour size less than 5 mm.120 Damian et al in largest series of 100 patients demonstrated a sensitivity of 93%. They also showed that FDG PET detected disease up to 6 months earlier than conventional imaging modalities and altered management in 22% patients.119 FDG PET can miss small tumours and micrometastases, in such cases sentinel lymph node biopsy using gamma probe is ideal procedure, which has sensitivity of more than 94% and specificity of 100%.121 FDG PET has better sensitivity of 94% and specificity of 94% as compared to CT, which has 55% and 84% respectively in detecting metastatic melanoma.122,123 FDG PET is more accurate than conventional imaging in restaging and follow-up of patients with melanoma, it has sensitivity of 92% vs 57.7% and a specificity of 94% vs 45% respectively.122
FDG-PET has been approved for the diagnosis, initial staging and restaging of malignant melanoma patients. However, evaluation of regional lymph nodes is still not covered, since sentinel lymph node detection is the method of choice for this purpose.
Gastric and Esophageal Cancer
FDG PET is sensitive for detection of primary gastric/esophageal squamous cell as well as adenocarcinoma.124-128 Although, the detection of primary tumour has been reported to be excellent, the identification of regional nodal metastases has been restricted due to involvement of small volume of disease in some of lymph node. But still FDG is more specific than CT and endoscopic US for locoregional lymph node metastasis.124-127 However, biopsy remains the gold standard for detecting lympg node involvement. FDG PET changed the management in 22% cases.124,125,127 Therefore PET is reliable in differentiating resectable from non-resectable disease and avoids many unnecessary surgeries (Figure 9).
FDG PET is more accurate than conventional imaging with CT and US in evaluating response to radiation therapy and chemotherapy. A recent study showed FDG PET is very sensitive in identifying responders to neoadjuvant therapy.129 The responders have significant change in FDG uptake as compared to non-responders after treatment.
FDG-PET has been recently approved for the diagnosis, initial staging and restaging of esophageal cancer. Also there are some encouraging reports regarding the importance of FDG-PET for therapy response monitoring in esophageal cancer. As these studies increase in number, PET will probably be used more in routine practice.
Multiple myeloma and osteosarcoma are most common malignant primary bone tumours. In multiple myeloma bone scan shows marrow involvement indirectly by imaging cortical bone osteoblastic reaction to the marrow tumour while FDG PET shows marrow involvement directly. Jadvar et al concluded that FDG PET could detect early marrow involvement as compared to bone scan and other conventional imaging. PET is also useful in assessing response to treatment.130 The other studies in this regard also showed encouraging results in demonstrating the active site of tumour involvement in planning the mode of treatment.131-133
In osteosarcoma, degree of FDG uptake in tumour has good correlation with histological grading or tumour aggressiveness.134,135 Eary et al reported that baseline SUV can predict out come in a better way than conventional grading in meta-analysis of 209 patients. FDG PET scan also guides biopsy to the most active tumor metabolic site.136 In osteosarcoma, since lymph node metastases are very rare the experience in detecting lymph node metastases is limited. Lucas et al showed that FDG detected more lesions as compared to CT and MRI.137 Recently, Brenner et al in their review of PET studies in osteosarcoma concluded that FDG PET is useful in predicting outcome as well as tumor response in neoadjuvant chemotherapy, differentiating postoperative changes from residual tumor tissue or local relapse and for detecting distant metastases.
The use of FDG-PET for bone tumors is not a Medicare approved oncologic indication yet.
Ovarian carcinoma is spread through out pelvis in about two third of all patients at the time of presentation. Accurate staging is very important for proper management, especially in patients with raised serum CA-125. FDG PET has shown high sensitivity and specificity as compared to conventional imaging modalities in detecting lymph node metastases, however like other imaging modalities, it has limited role in micrometastases and very small lesions (Figure 10).139-141 Surgical biopsy staging remains gold standard to know small lymph nodes metastases and micrometastases.
Although FDG-PET has not yet been approved by the Health Care Finance Administration for Medicare reimbursement for ovarian cancer, it may be used depending on the patient's insurance. More studies regarding the importance and necessity of PET imaging for ovarian cancer is still warranted.
Cancer of Unknown Primary (CUP)
Many studies have been published in the literature to evaluate the role of PET in patients with cancer of unknown primary.142-144 However, there are very few studies to date, which have analyzed the impact of PET results on therapeutic management. Rades et al detected the primary site in 18 of 42 patients using PET, all these patients had localized cancer of unknown origin by using conventional staging procedures.145 PET was positive for disseminated diseases in 38% patients. In 69% patients PET results influenced the selection of the definitive treatment.
Positron emission tomography is an advanced diagnostic functional imaging technique, which can not only detect but also quantify disease processes in vivo non-invasively. The principle of PET is to detect altered metabolism of disease processes and not the altered anatomy. 18F-FDG is the most widely used radiotracer in oncology. Since glucose metabolism is increased many fold in malignant tumors as compared to normal cells, PET has high sensitivity and high negative predictive value. It has a well-established role in initial staging, monitoring response to the therapy, and management of many types of cancer, including lung cancer, colon cancer, lymphoma, melanoma, esophageal cancer, head and neck cancer, and breast cancer.
We are thankful to International Union Against Cancer (UICC), Geneva, Switzerland for financial support (ACSBI fellowship) to Dr. Rakesh Kumar.
Copyright 2003 - Indian Journal of Cancer
The following images related to this document are available:
Photo images[cn03015f9.jpg] [cn03015f8.jpg] [cn03015f3.jpg] [cn03015f2.jpg] [cn03015t1.jpg] [cn03015f1.jpg] [cn03015f6.jpg] [cn03015f4.jpg] [cn03015f5.jpg] [cn03015f10.jpg] [cn03015f7.jpg]