How we read FCH-PET/CT for prostate cancer
© The Author(s). 2016
Received: 15 October 2016
Accepted: 1 December 2016
Published: 6 December 2016
Over the last decade, 18F-fluorocholine positron emission tomography/computed tomography (FCH-PET/CT) has gained in popularity for the staging and restaging of patients with prostate cancer (PCa). However, despite abundant literature on the topic, there is a lack of publications on how to actually interpret FCH-PET/CT in a clinical setting. Here we propose a practical, TNM-oriented approach to read FCH-PET/CT, with notes on procedure technique, image display, review sequence and report structure. The purpose of this article is to provide guidance to radiologists, nuclear medicine physicians and residents who are new to FCH-PET/CT, as well as to propose an alternate approach to more experienced physicians.
Positron emission tomography/computed tomography (PET/CT) with radiolabeled choline, a marker of cell membrane synthesis, has proven to be a useful imaging technique in the management of prostate cancer (PCa). 18F-fluorocholine (FCH) is more convenient than 11C-choline, the tracer of reference, as it can be delivered to centers devoid of an on-site cyclotron, owing to the longer half-life of 18F. 11C-choline is now FDA-approved for restaging of PCa in the setting of biochemical relapse after radical prostatectomy , and this indication is also the most established one for FCH-PET/CT to date [2–4]. FCH-PET/CT is also useful for initial stating of patients with high-risk PCa [2, 4]. FCH usually refers to 18F-fluoromethylcholine, but 18F-fluoroethylcholine has also been employed. Both tracers appear to have a comparable biosdistribution , although 18F-fluoromethylcholine has been used and studied more extensively . The reading approach presented here applies to both tracers.
Contrary to 18F-fluorodeoxyglucose (FDG)-PET/CT, fasting prior to FCH-PET/CT is not required, as FCH biodistribution and tumor uptake are not affected by glycemia or insulinemia. Androgen deprivation therapy (ADT) does not need to be stopped before FCH-PET/CT performed in the setting of biochemical recurrence under ADT, as castration-resistant PCa lesions are proliferating despite ADT .
Tumor uptake and blood clearance of FCH are rapid. Because of this a prolonged uptake period, such as 60 min when doing an FDG-PET/CT, is not mandatory with FCH. At our center, FCH is rather injected on the scanner table while simultaneously starting a 10-min dynamic acquisition over the pelvis. This allows visualization of early pathological uptake in the prostate or the prostatic bed, as well as in the regional nodal stations, before the urine activity enters the distal ureters and bladder, thus avoiding the need for a delayed or post-diuretic acquisition, or bladder catheterism. The dynamic pelvic acquisition can be reconstructed in fixed (e.g. 5 × 2 min) or progressively longer frames (e.g. 4 × 30 s, 4 × 1 min and 2 × 2 min).
The dynamic acquisition is immediately followed by the whole-body (WB) acquisition from skull vertex to proximal thighs, with arms preferably upward. It is rarely necessary to include the distal upper or lower limbs, unless there are known or suspected metastases beyond proximal part thereof. Contrary to FDG-PET/CT however, skull and brain are systematically included in FCH-PET/CT acquisition because (1) skull is a frequent site of axial bone metastases in PCa, and (2) brain is much easier to assess than with FDG-PET/CT owing to the lack of physiological FCH uptake in the normal brain. Oral contrast can optionally be administered to better discriminate intestinal loops and confirm physiological uptake or pancreatic excretion of FCH along the gastrointestinal tract.
PET color maps and fusion blend
The PET-only images, including MIP, are displayed with an inverse linear gray look-up table (LUT) color map. When fused with CT, a rainbow LUT (e.g. “ECAT Rainbow” found on Siemens’ systems; Fig. 1) is preferred, as it provides more contrast over a wider dynamic range than LUTs incorporating fewer colors (such as red and yellow “Hot Metal” LUT). A rainbow LUT starting with black naturally attenuates the unwanted low-level background noise. The upper end of the LUT should preferably not be white, but rather of a bright color such as red, in order to minimize the interference with the underlying black and white CT signal. Furthermore, a desired feature of the chosen rainbow LUT is a logarithmic progression, which naturally allows distinguishing uptake level differences that are clinically meaningful, both in the lower and the higher ends of the uptake intensity spectrum. A rainbow LUT with these characteristics facilitates performing most of the orthogonal slices review in fusion mode. A linear blending mode with an alpha blend of 50% should be used in fusion mode (Fig. 1), so that both CT and PET images are equally represented and neither modality is ever masked by the other, particularly in areas of high uptake or density.
PET images normalization and visual uptake assessment
Unlike FDG, FCH has a relatively high physiological uptake in the liver. Because of this, FCH images are normalized by making the liver appear yellow to orange, on average, and not saturated (Fig. 1). We find that adjusting the inverse gray scale LUT of PET-only images to the same upper standardized uptake value (SUV) threshold as the color LUT provides an adequate normalization thereof, with the same dynamic range as color PET images. Once the whole-body PET is properly normalized, the dynamic pelvic PET images (which do include the liver) are normalized to the same upper SUV threshold as that of the WB PET images.
Proposed FCH visual uptake scale for PCa lesions
(> ~ 1 cm)
(≤ ~1 cm)
Not visible (equal to or lower than background)
Higher than background, but lower than that of the bone marrow
Similar to or higher than that of the bone marrow, but lower than that of the liver and exocrine glandsb
Similar to or between that of the liver and exocrine glands
Higher than that of the liver and exocrine glands
Comparable to or higher than that of the kidney
Very intense uptake
Red (large area of saturation)
Normal FCH biodistribution
Before attempting to interpret FCH-PET/CT, one must be familiar with the normal biodistribution and kinetics of the tracer, which has been covered in details elsewhere . On the dynamic scan, the first-pass vascular bolus is noted in the iliac arteries on the first frame(s) and rapidly fades away. Urine activity typically reaches distal ureters and the bladder at around 4 to 6 min, but sometimes later.
Kidneys have a very intense physiological uptake, usually the highest among healthy organs, and are almost always saturated when PET images are normalized as above. Liver uptake is quite intense, but less than that of the kidneys. The uptake in spleen and major exocrine glands (pancreas, salivary and lachrymal glands) is usually moderate and lower than that of the liver, but sometimes comparable or slightly above the latter. Bone marrow and endocrine glands (pituitary, thyroid, adrenals) generally show faint to mild uptake. Uptake in the gastro-intestinal tract is highly variable in extent and intensity, and is related to the rapid turnover of mucosal cells, in addition to probable excretion of FCH within pancreatic juices (hypothesized from our frequent observation of continuous activity extending from the duodenum along downstream small bowel). Skeletal and cardiac muscles, as well as choroid plexus exhibit faint and diffuse uptake. Marked uptake is frequently seen along the vein proximal to injection site (Fig. 1). The normal prostate exhibits minimal uptake.
The TNM approach
A problem-oriented approach to reading FCH-PET/CT (and oncological PET/CT in general) is better achieved by adopting a tumor-node-metastasis (TNM) sequence than using the typical radiological anatomical segmentation (i.e. head and neck, thorax, abdomen and pelvis). Indeed, PET/CT being a systemic imaging primarily used to rule in or out a systemic disease (i.e. the presence or not of metastasis), compartments delimited by anatomic boundaries such as the thoracic inlet or the diaphragm make little sense when describing the extent of PCa disease. Indeed, beginning with description of findings in head and neck area, while PCa actually originates from the pelvis and spreads upward, is counter-intuitive. Furthermore, bone, the most frequent site of distant metastasis in PCa, is an organ present in all radiological segments and is more logically subdivided into axial and appendicular skeleton in the oncological setting. In fact, bone scan is seldom interpreted using conventional cross-sectional anatomical divisions, as bone metastases do not recognize those arbitrary boundaries. The same logic applies to FCH-PET/CT.
Accordingly, in the context of PCa, we begin by describing findings in prostate or prostatic bed, then those in proximal to distal nodal stations relative to the prostate, and finally those at potential sites of distant metastases starting with the most frequent, the skeleton. For each TNM station, a clear and concise description of PCa-relevant positive and negative findings is warranted. This is followed by description of incidental findings, if any, finishing with accessory findings.
Image review order
Although the dynamic acquisition is performed first, we find it more efficient to begin by reviewing the WB scan. The first series to look at when opening the FCH-PET/CT study is the WB AC-PET MIP, which is the most informative and fastest to assess PET-only representation. In just a few seconds, is it possible in the vast majority of cases to determine the M and even N statuses on the MIP. Even the subtlest foci of uptake can be seen in at least a few projection angles, and can then easily be located onto the orthogonal views with a simple mouse click. When distant metastases are present (M1), the patient is generally inoperable. Having this information in mind from the start will allow the reader to be more concise and clinically relevant when describing the loco-regional findings. Contrary to conventional cross-sectional imaging, PET MIP allows looking at the big picture in a trice, i.e. seeing the forest for the trees. In a patient with numerous metastases, the emphasis will be on that big picture (e.g. overall tumor burden, sites involved, metabolic intensity and heterogeneity, etc.), rather than on minute loco-regional details or exact number of lesions, which would be more important to emphasize in a surgical case.
Following the general survey based on MIP, we proceed with a careful review of PET/CT fused transaxial slices (referring to coronal and sagittal slices as needed). CT-only images are consulted as needed for better locating, delineating or measuring lesions of interest. Throughout reading, CT windowing is alternated between soft tissue, bone, lung and brain windows as appropriate. We find that orthogonal AC PET-only slices are less often consulted because rotating MIP already provides an excellent display of PET alone in a very practical format. However, PET-only slices are sometimes helpful to distinguish between what constitutes a definite focus of mild uptake that may suggest the presence of a small lesion, versus non-specific heterogeneity or statistical noise.
NAC PET images are useful for quality control purposes, and to assess areas or abnormalities prone to attenuation correction artifacts and misregistration. Review of lung fields on NAC PET, where they are naturally overrepresented, can improve the sensitivity to detect or confirm small foci of uptake. We therefore tend to review NAC images at the time of surveying lungs for metastases. After going through the TNM sequence, as described below, there will be a rapid, general survey of CT-only transaxial slices, looking for potential incidental or accessory findings not already identified on PET and fused images.
T: Local compartment assessment
N: Regional compartment assessment
Contrary to conventional imaging, there is no size criterion for positivity in PET/CT. The lack of uptake in a node >10 mm makes the latter more likely benign. On the other hand, PET is frequently frankly positive for nodes <10 mm.
M: Distant compartment assessment
The first organ to assess here is the skeleton, which is by far the commonest site of distant metastasis in PCa. Typically, bone metastases from PCa are sclerotic, but it is not uncommon that lesions are initially lytic, particularly in untreated or high-grade PCa. FCH uptake is variable in bone metastases and has been shown to inversely correlate with bone density . This is consistent with sclerotic lesions being generally less aggressive, of lower tumor cell density, and/or having responded to therapy. Contrary to most other cancers, spread of PCa to the bone can be regional via the venous plexus of Batson, resulting in bone metastases being more frequent in the pelvis and lower spine. In addition to the location, number (approximate when numerous) and uptake features of suspicious lesions, we also comment on their radiological appearance (lytic, sclerotic or mixed), on the presence of metabolic and/or radiological heterogeneity among lesions, as well as on the risk of pathological fracture (with lesions in weight-bearing bones, of lytic appearance, and/or involving more than 2/3 of bone width harboring the highest risk ) or other complications such as cord involvement or compression. Additionally, the presence of bony lesions on CT that are not enhancing and are thought to represent benign lesions (such as benign bone islands) or healed metastases should be mentioned, as they could potentially raise suspicion on other radiological exams. Of note, the sagittal view is particularly convenient to assess the spine, as it allows a complete survey thereof over a very few slices. In addition, vertebral fractures are also easily detected in this view.
Other potential sites where metastases needs to be ruled out include the liver, adrenals, lungs and brain, the latter being easier to assess with FCH- than with FDG-PET/CT, as mentioned earlier. The sensitivity of FCH-PET/CT for small liver metastases is reduced, as they either need to exhibit a level of uptake that is superior to that of the liver, or to be large enough so that significant uptake can be evidenced with confidence and despite spill-in from surrounding liver activity. Because liver metastases are relatively infrequent in PCa, this remains a relatively minor drawback of FCH-PET/CT.
If there is no evidence of metastasis, a simple phrase such as “There are no foci of increased uptake elsewhere that are suspicious for metastases, in particular in the bones” will concisely summarize relevant negative findings. Because at our institution CT performed as part of a PET/CT procedure is acquired at low amperage, without IV contrast and in free-breathing conditions, our focus is not on trying to find – or exclude the presence of – any small lesion without detectable FCH uptake, which could otherwise be better visualized on a fully diagnostic CT or MRI scan. Considering that the negative predictive value of FCH-PET/CT has been shown to be superior to that of conventional imaging [2, 3], such small lesions without detectable FCH uptake are less likely to be clinically relevant.
Incidental and accessory findings
An incidental finding is a condition unrelated to that for which the FCH-PET/CT was requested, and which presence was unknown before the procedure. It differs from an accessory finding in that it may potentially alter patient management. FCH uptake is not specific to PCa, and most cancers other that PCa can also exhibit FCH uptake. An FCH-avid lesion in a location that is atypical for PCa spread – particularly in the absence of more typically located metastases elsewhere – should raise the possibility of a synchronous malignancy. Depending on the context of the patient, further investigation of such an incidental finding may or may not be clinically relevant. Other incidental findings that are more easily detected on CT (even on low-dose CT) include potentially life-threatening conditions such as a large vascular aneurysm, a pneumothorax or a cavity effusion.
Accessory findings are abnormalities of lesser clinical interest to the referring physician, which do not fall under the PCa-relevant positive or negative findings, or incidentalomas categories. Virtually any benign lesions, traumatic or inflammatory processes that can demonstrate hypermetabolism with FDG can also exhibit increased FCH uptake to a variable extent. Again and again, it is the distribution pattern of hyperactive lesions that is the most determinant factor in the diagnostic decision-making process. Uptake intensity may modulate the level of suspicion of malignancy, but is not diagnostic by itself. CT appearance is often helpful to increase the level of confidence. For example, an FCH-avid adrenal nodule will confidently be classified as a benign adenoma in the presence of a low attenuation on non-contrast low-dose CT. Other examples of accessory findings include FCH uptake related to osteoarthritis, hilar and mediastinal reactive lymph nodes, reactive bone marrow (diffuse uptake), and traumatic lesions. Accessory findings detected on CT only, i.e. without metabolic correlate, include surgical sequella unrelated to PCa (of which cholecystectomy is probably be the most common), vascular calcifications, steatosis, renal or liver cysts, and old fractures, just to name a few.
A normal structure with an FCH uptake within normal range should not be described. This only lengthens the report, without adding clinical value. Finally, for the reasons mentioned earlier concerning low-dose CT, we do not describe any organ as having a “normal” appearance on CT, as a fully diagnostic CT could contradict this.
Tumor SUVs are not systematically reported because they have not consistently been shown to improve diagnostic accuracy over qualitative interpretation. There is no clinically proven FCH SUV cutoff that can rule in or out malignancy. SUV is heavily influenced by the partial volume effect. PET can detect small, even sub-cm metastases, but SUV will inevitably be lower in such lesions. Visual assessment may, at least partly, overcome this limitation (Table 1).
SUVs are useful in research as a continuous quantitative variable to be correlated with tumor or patient characteristics or outcomes in a given population. But, in the clinical setting, the only utility of SUV is to compare uptake values between repeat studies in the same patient, e.g. for therapeutic response assessment. Even then, SUV comparison remains in support of qualitative uptake assessment, and is valid only if the current and prior studies are normalized to a similar SUV upper threshold, implying that the organ of reference’s uptake is stable. While a SUV difference of 30% is generally considered significant, i.e. unlikely due to chance , it is our experience that such a difference is easily picked up visually when images are properly normalized. Furthermore, complete molecular response remains a visual criterion, as SUV can virtually never reach zero. Hence, SUV comparisons have a limited added value other than illustrating the amplitude of the lesion’s partial response or progression.
In the end, in the oncologic setting, the foremost important feature of scintigraphic abnormalities, before considering any uptake quantification, is their distribution pattern relative to each other and to the primary tumor. A high-uptake lesion in an uncommon location for a metastasis is most often not one, particularly if no other more typically located metastases are found. Likewise, a definite focus of mild uptake in a rounded sub-cm lymph node in the obturator area ipsilateral to the primary PCa lesion will raise some suspicion of metastasis, regardless of the actual SUV. In both situations, PCa-specific pattern recognition and clinical judgment are key to accurate diagnosis, not SUVs. In our experience, consistent qualitative appreciation of uptake intensity (e.g. as in Table 1) instead of systematically reporting SUVs has many advantages: (1) it avoids pitfalls related to SUV measurements, (2) it minimizes the too often overstated importance of SUVs and (3) it makes the report more concise.
Our reports are divided in 4 sections, as follows:
Clinical information section
Clinical information available at the time of reporting should be summarized. This includes the relevant information gathered from the requisition form (including the clinical question to be answered), the patient questionnaire and interview, the medical record, or the treating team. Details about the pathology results (including Gleason score), prior and current treatments, other imaging findings, biochemical status (PSA and, if relevant, PSA doubling time) and symptoms that are relevant for the interpretation of the FCH-PET/CT scan should be mentioned.
This section is brief and limited to few sentences. It is not necessary to mention every technical detail that can be found elsewhere, in the standard procedure protocol or patient imaging record, unless there has been a significant protocol deviation that could potentially impact on the quality of the scan. We favor a brief statement about the imaging sequence and technique, the areas surveyed, and the use of contrast or drug if any. For example: “At the time of FCH administration, we started a 10-min dynamic PET acquisition over the pelvis, which was followed by a whole-body acquisition from vertex to proximal thighs. Both acquisitions were accompanied by a low-dose CT with oral contrast for anatomical correlation and attenuation correction.”
The observation section describes the relevant positive and negative findings in a TNM sequence, as detailed earlier. For simpler cases, one paragraph is generally sufficient, while for more complex cases T, N and M compartments can be separated in distinct paragraphs. The emphasis is on the description of the findings, rather than opinions.
If there were any incidental finding, its description would come thereafter, in a separate paragraph. This is followed by description of accessory findings, starting with “Accessorily, …”. Because accessory findings are most of the time benign and unrelated to the oncological condition, their description should be succinct. Further, for these we tend to formulate a diagnostic opinion together with the description within the Observations section, particularly for obviously benign findings that would be superfluous to point out again in the conclusion. An example of this could be: “There is mild uptake in multiple hilar and mediastinal lymph nodes, which are symmetric in distribution and thus likely of inflammatory or granulomatous etiology.”
In the case of a repeat study, to assess either the evolution under observation or the therapeutic response, the description should focus on the metabolic (and secondarily anatomic) evolution of previously seen PCa lesions, as well as the apparition of new suspicious lesions if any. We would also comment on the homogeneity or heterogeneity of lesions’ evolution or response. Here there is no need to mention again all accessory findings if they are stable and clinically irrelevant. A simple statement such as “The remainder of the scan shows the same accessory findings as previously described. No other significant changes are seen as compared to the previous study” can be used instead.
This is the most important section of the report, and in the majority of cases, the only one the referring physician will ever read. Conclusion should straightforwardly answer the clinical question, which is typically to stage or restage PCa. Hence, the first sentence(s) should formulate as concisely as possible an opinion on what the TNM stage is. Completely equivocal opinions should be avoided as much as possible. In those more difficult cases, the arguments for and against a given hypothesis can be summarized in the conclusion, ending up with a likelihood statement about the hypothesis. These arguments may include clinical information affecting the probability of a positive (or negative) finding being truly positive (or negative). For example, it has been demonstrated that a high Gleason score, a high PSA and a short PSA doubling time are associated with higher sensitivity of FCH-PET/CT . Hence, an equivocal finding based on FCH-PET/CT images alone, such as a solitary pelvic lymph with a rather mild FCH uptake, could be interpreted as more likely to represent metastasis in a patient with any of these characteristics, or less likely to in a patient with none.
Opinion on the probable nature of any incidental finding is given thereafter. However, accessory findings are only mentioned in the conclusion if they are likely to alter patient care. For example, we seldom mention the presence of cholecystectomy sequela in the Conclusion section of an oncological PET/CT report.
The findings are consistent with a primary prostate cancer lesion extending over the middle and apical sextants of the left prostatic lobe, accompanied by two nodal metastases in the left external iliac area.
No evidence of osseous or visceral metastases.
Incidental finding of a thyroid nodule with moderate uptake in the right thyroid lobe, for which a primary thyroid malignancy cannot be ruled out. This could further be investigated with ultrasound and fine-needle biopsy if clinically relevant in the setting of this patient.
The findings are consistent with a complete metabolic response.
In summary, FCH-PET/CT for PCa is read and reported in a similar fashion as oncological FDG-PET/CT. For the experienced FDG-PET/CT reader, the learning curve is rapid, and involves mastering the normal biodistribution of FCH, the dissemination pattern of PCa, and benign pathologies exhibiting FCH uptake (largely overlapping with those associated with FDG uptake). Reviewing the dynamic PET images may be challenging at first, as not all software viewers are user-friendly for this purpose. However, there are overall more similarities than differences between reporting oncological FCH- and FDG-PET/CT. For both, we encourage adopting a problem-oriented approach based on the TNM system, in combination with a concise reporting style clearly addressing the clinical question upfront, both in the Observations and Conclusion sections. In our opinion, this makes PET/CT reading more efficient for the nuclear medicine physician and radiologist, and report clearer and more clinically relevant for the referring physician or surgeon.
J.M.B. is supported by a Clinical Research Scholar grant from the Fonds de recherche du Québec – Santé.
Availability of data and materials
The authors declare that they have no competing interests.
J.M.B. wrote the manuscript. A.B. composed the figures and revised the manuscript. All authors read and approved the final manuscript.
Consent for publication
Ethics approval and consent to participate
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Anonymous. FDA approves 11C-choline for PET in prostate cancer [Newsline]. J Nucl Med. 2012;53(12):11N. http://jnm.snmjournals.org/content/53/12/11N.full.pdf+html.
- Vali R, Loidl W, Pirich C, Langesteger W, Beheshti M. Imaging of prostate cancer with PET/CT using 18F-Fluorocholine. Am J Nucl Med Mol Imaging. 2015;5(2):96–108.PubMedPubMed CentralGoogle Scholar
- Beauregard J-M, Pouliot F. New developments in the imaging of metastatic prostate cancer. Curr Opin Support Palliat Care. 2014;8(3):265–70.View ArticlePubMedGoogle Scholar
- Bauman G, Belhocine T, Kovacs M, Ward A, Beheshti M, Rachinsky I. 18F-fluorocholine for prostate cancer imaging: a systematic review of the literature. Prostate Cancer Prostatic Dis. 2012;15(1):45–55.View ArticlePubMedGoogle Scholar
- Filice A, Asti M, Roncali M, Fraternali A, Casali M, Di Paolo M, Guidotti C, Versari A, Salvo D. Comparison between the physiological distribution of 18F-FMCH and 18F-FECH. Eur J Nucl Med Mol Imaging. 2010;37(S2):S426.Google Scholar
- Evangelista L, Cervino AR, Guttilla A, Zattoni F, Cuccurullo V, Mansi L. 18F-fluoromethylcholine or 18F-fluoroethylcholine pet for prostate cancer imaging: which is better? A literature revision. Nucl Med Biol. 2015;42(4):340–8.View ArticlePubMedGoogle Scholar
- Giovacchini G, Breeuwsma AJ. Restaging prostate cancer patients with biochemical failure with PET/CT and radiolabeled choline. Q J Nucl Med Mol Imaging. 2012;56(4):354–66.PubMedGoogle Scholar
- Beauregard J-M, Giraudet A-L, Aide N, Hofman MS, Blum R, Drummond E, Roselt P, Hicks RJ. Evaluation of a new visual uptake scoring scale for 18F-fluorothymidine positron emission tomography in the diagnosis of pulmonary lesions. Nucl Med Commun. 2013;34(6):521–6.View ArticlePubMedGoogle Scholar
- Beheshti M, Haroon A, Bomanji JB, Langsteger W. Fluorocholine PET/computed tomography: physiologic uptake, benign findings, and pitfalls. PET Clin. 2014;9(3):299–306.View ArticlePubMedGoogle Scholar
- Presti JC. Prostate biopsy: current status and limitations. Rev Urol. 2007;9(3):93–8.PubMedPubMed CentralGoogle Scholar
- Beheshti M, Vali R, Waldenberger P, Fitz F, Nader M, Hammer J, Loidl W, Pirich C, Fogelman I, Langsteger W. The Use of F-18 Choline PET in the Assessment of Bone Metastases in Prostate Cancer: Correlation with Morphological Changes on CT. Mol Imaging Biol. 2009;12(1):98–107.View ArticlePubMedGoogle Scholar
- Jawad MU, Scully SP. In Brief: Classifications in Brief: Mirels’ Classification: Metastatic Disease in Long Bones and Impending Pathologic Fracture. Clin Orthop Relat Res. 2010;468(10):2825–7.View ArticlePubMedPubMed CentralGoogle Scholar
- Wahl RL, Jacene H, Kasamon Y, Lodge MA. From RECIST to PERCIST: Evolving Considerations for PET response criteria in solid tumors. J Nucl Med. 2009;50 Suppl 1:122S–50S.View ArticlePubMedPubMed CentralGoogle Scholar