- Email: [email protected]
CT in anal cancer — is it worth doing?
Clinical Radiology 67 (2012) 535e540
Contents lists available at SciVerse ScienceDirect
Clinical Radiology journal homepage: www.clinicalradiologyonline.net
PET/CT in anal cancer d is it worth doing? I.T. Wells*, B.M. Fox Department of Radiology, Derriford Hospital, Plymouth, Devon, UK
art icl e i nformat ion Article history: Received 29 July 2011 Received in revised form 7 October 2011 Accepted 12 October 2011
AIM: To evaluate the role of 2-[18F]-fluoro-2-deoxy-D-glucose (FDG) positron-emission tomography (PET)/ computed tomography (CT) in the current multidisciplinary management of anal cancer, both in initial staging and in follow-up post-treatment. MATERIALS AND METHODS: All patients referred to the region-wide multidisciplinary meeting for anal cancer during the study period received PET/CT imaging in addition to conventional imaging [CT and magnetic resonance imaging (MRI)]. Whether PET/CT altered the stage of the tumour from that suggested by conventional imaging was retrospectively assessed. The effect on management was evaluated. RESULTS: Fifty PET/CT examinations were performed on 44 patients with anal cancer. Thirty were part of initial staging, and 20 were post-chemo/radiotherapy or surgery. Two PET/CTs produced inadequate contemporaneous conventional imaging to allow comparison. Overall PET/CT increased the stage of the anal cancer in 17% of cases (8/48), decreased the stage in 19% (9/48), and did not alter the stage in 65% (31/48). The tumour stage was altered more frequently in initial staging than in follow up imaging. The PET/CT findings altered patient management in 29% (14/48) of cases. The majority (11) of these were cases in which PET/CT was used as part of initial staging. CONCLUSION: PET/CT alters the initial staging sufficiently frequently that it should be used routinely in anal cancer, where it is available. The role of PET/CT in the follow-up of anal cancer is not as clear. Routine follow-up with PET/CT may not be justified, but selected use is of definite benefit in problem solving or if salvage surgery is planned, after multidisciplinary discussion. Ó 2011 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
Introduction Anal cancer is rare, with an incidence of 1e2 cases per 100,000 in the UK, although this is believed to be increasing with increased exposure to human immunodeficiency virus and human papilloma virus.1 Limited data exist on the best imaging strategy for the initial staging and follow-up of anal cancer. Several small series (the largest by Winton et al. consisting of 61 examinations, in the Peter MacCallum * Guarantor and correspondent: I.T. Wells, Department of Radiology, Derriford Hospital, Plymouth, Devon PL6 8DH, UK. Tel.: þ44 (0) 1752437437; fax: þ44 (0) 1752315300. E-mail address: [email protected] (I.T. Wells).
Cancer Centre Australia)2 have shown that 2-[18F]-fluoro2-deoxy-D-glucose positron-emission tomography (FDG PET) alters the initial staging of anal cancer in a significant number of patients. A few small retrospective series comment on its role in the follow-up of anal cancer patients. In the southwest of England, cases from five sites are discussed at a monthly anal cancer multidisciplinary team (MDT) meeting. This has provided the opportunity to evaluate the role of FDG PET/computed tomography (CT) in current multidisciplinary management of anal cancer, both in initial staging and in follow-up post-treatment. This is particularly relevant at a time when financial resources are stretched, as PET/CT remains an expensive investigation. The experiences in the southwest of England are also widely
0009-9260/$ e see front matter Ó 2011 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.crad.2011.10.030
536
I.T. Wells, B.M. Fox / Clinical Radiology 67 (2012) 535e540
applicable; although patients are discussed at a region-wide MDT meeting, they are imaged and managed at their local hospitals. Through reviewing the MDT discussion process, the present authors have been able to evaluate not only the effect of PET/CT on cancer stage, but on actual patient management. It also allows some reflection on the usefulness of the MDT discussion process, when increasing demands on the time of radiologists, pathologists, oncologists, and surgeons are putting strain on the continued commitment to this model of patient care.
proceed to chemo/radiotherapy or operation, changed the radiotherapy field by detecting unsuspected nodal involvement, or changed a plan to resect liver or lung metastases (where this was previously being considered). Also recorded was the frequency of the primary anal tumour being identified on PET/CT, both on initial staging images and on follow-up images. Extra investigations suggested by the PET/CT images for incidental findings were also documented to allow a qualitative assessment of the financial impact of this.
Results Materials and methods Local review deemed that full ethics approval was not required as the nature of the study is a retrospective audit. All PET/CT examinations for patients with anal cancer performed in the southwest of England since the service started in 2008 were identified by performing a radiology information systems search, and cross-referencing this with records from the anal cancer MDT and the personal records kept by the radiology consultants reporting PET/CT. For each patient identified, the conventional imaging was reviewed and TNM classification and stage of the anal tumour retrospectively recorded. This was done primarily by reading the reports written at the time, with re-review of the images only when the reports were unclear. Conventional imaging consisted of CT (thorax abdomen, and pelvis with intravenous contrast media, and thin-section multiplanar reformat), plus MRI pelvis (small field of view axial, coronal and sagittal T2, large field of view T1 and T2). This region-wide common approach was agreed through the MDT. In all the cases included in the study, conventional imaging was performed within 6 weeks of the PET/CT. If the time gap was larger, or if there was any significant change in therapy between the conventional imaging and the PET/CT being performed, the case was not used. Staging was performed using the sixth edition of the American Joint Committee on Cancer guidelines. The reports of the PET/CTs were then reviewed and TNM classification used to asses tumour status. This was recorded on a separate spreadsheet. Whether the patients were pre or post-initial treatment at the time of the examination was recorded. All PET/CT examinations were performed on a Philips Gemini portable scanner (Philips Medical Systems, Cleveland, OH, USA) after intravenous injection of FDG. CT was obtained without intravenous contrast medium and used for attenuation correction. The frequency with which PET/CT altered the stage of the tumour from the stage suggested by conventional imaging was assessed. These figures were analysed overall and after stratification by initial T stage, N stage and M stage. The subsequent management of the patients was evaluated by looking through subsequent imaging, request cards, notes from the anal cancer MDT meeting, and patient notes where further information was required. The management was considered to be changed if the PET/CT altered a plan to
During the period from June 2008 to May 2011, 50 PET/CT examinations were performed on 44 different patients in the southwest of England with anal cancer. Thirty of these were part of initial staging, and 20 were post-chemo/ radiotherapy or surgery. The follow-up ranged from 6 weeks to 2 years post-treatment. Two PET/CTs produced inadequate contemporaneous conventional imaging to allow comparison. The results are shown in Table 1. Overall PET/CT increased the stage of the anal cancer in 17% of cases (8/48), decreased the stage in 19% (9/48), and did not alter the stage in 65% (31/48). If only the cases in which the PET/CT was performed for initial staging are included, PET/CT increased the overall stage of the anal cancer in 20% of cases (6/30), decreased the stage in 27% (8/30), and did not alter the stage in 53% (16/30). If only the cases in which the PET/CT was performed for follow-up are analysed, PET/CT increased the overall stage in 11% of cases (2/18), decreased the stage in 6% (1/18), and did not alter the stage in 83% (15/18). If the data are broken down by T-stage, the overall stage was changed (either up or down) in 20% (4/20) of T1 tumours, 45% (5/11) of T2 tumours, and 47% (7/15) of T3 tumours. Only one Tx and one T4 tumours were scanned, and the stages of these were unchanged by PET/CT. This is presented in Table 2. If the data are broken down by N-stage, the nodal stage was changed (either up or down) in 12% (3/26) of N0 tumours, 50% (3/6) of N1 tumours, 50% (6/ 12) of N2, and 0% (0/4) of N3 tumours. This is presented in Table 3. Overall, the N stage was altered by PET/CT in 25% (12/48) of cases. If the data are broken down by M-stage, the metastatic stage was upstaged in 18% (6/33) of M0 tumours, and downstaged in 27% (4/15) of M1 tumours. This is presented in Table 4. Overall, the M stage was altered by PET/CT in 21% (10/48) of cases. Table 1 Summary of the effects of positron-emission tomography (PET)/computed tomography (CT) on anal cancer stage compared to conventional imaging.
Overall, including all examinations Only including initial staging examinations Only including follow-up examinations
Upstaged by PET/CT
Downstaged by PET/CT
Unchanged by PET/CT
17% (8/48)
19% (9/48)
65% (31/48)
20% (6/30)
27% (8/30)
53% (16/30)
11% (2/18)
6% (1/18)
83% (15/18)
I.T. Wells, B.M. Fox / Clinical Radiology 67 (2012) 535e540 Table 2 Effect of positron-emission tomography (PET)/computed tomography (CT) on overall stage of anal tumour compared to conventional imaging, broken down by T stage.
Upstaged Downstaged Unchanged
Tx
T1
T2
T3
T4
Total T0e4
0 0 1
1 3 16
1 4 6
5 2 8
0 0 1
7 9 32
The findings of the PET/CT examinations altered patient management in 29% (14/48) of cases. The majority (11) of these were cases in which PET/CT was used as part of initial staging. The remaining three cases in which management was altered by the PET/CT were follow-up imaging after chemo/radiotherapy. Of the 14 cases where management was changed by the PET/CT findings: three were of patients with unsuspected distant nodal metastases found on PET/ CT, in the internal iliac, para aortic and aortocaval regions (Fig 1), and at the aortic bifurcation; one was a regional (inguinal) node not suspected on conventional imaging but with avid uptake on PET/CT; one was a case in which both new regional nodes and distant nodal metastases (paraortic) were found on PET/CT; one was an inguinal node not suspected on conventional imaging but with some metabolic activity on PET/CT (SUV 3.2), for which fine-needle aspiration (FNA) was performed showing this to be a reactive node; three were of lymph-node suspicious on conventional imaging but not metabolically active on PET/CT, which altered the radiotherapy planning field. Subsequent follow-up has confirmed non-involvement in two of these. It is too early to prove this in the other case; one was a case in which a liver metastasis suspected on conventional imaging was negative on PET/CT, but it picked up a new node altering the radiotherapy field (although histological proof of involvement could not be obtained); one was a case in which a liver metastasis though to be single on conventional imaging and planned for resection was found to be multiple on PET/CT (confirmed by MRI); three were of a metastasis suspected on CT but not metabolically active on PET/CT (two in the liver also assessed by MRI, one in the bone). Follow-up of greater than a year was available in two of these three cases to show that these were indeed not metastatic lesions. The primary tumour was seen to be metabolically active on PET/CT in 93% (28/30) of examinations performed as part of the initial staging. In the two initial staging examinations in which the primary was not identified prior excision biopsy had been performed. The primary tumour was evident on 56% (10/18) of the follow-up studies. Table 3 Effect of positron-emission tomography (PET)/computed tomography (CT) on nodal stage of anal tumour compared to conventional imaging, broken down by N stage.
Upstaged Downstaged Unchanged
N0
N1
N2
N3
Total N0e3
3 0 23
1 2 3
2 4 6
0 0 4
6 6 36
537
Table 4 Effect of positron-emission tomography (PET)/computed tomography (CT) on metastatic stage of anal tumour compared to conventional imaging, broken down by M stage.
Upstaged Downstaged Unchanged
M0
M1
Total M0e1
6 0 27
0 4 11
6 4 38
Eight further investigations were suggested in the reports of the PET/CTs: three upper GI endoscopies for uptake in the lower oesophageal sphincter, stomach, and duodenum (two performed with no abnormality identified), three colonoscopies for possible polyps (two performed with a polyp identified on one), one groin node FNA (negative), and one parotid ultrasound (normal).
Discussion This retrospective study adds to the limited data available on the use of PET/CT in the initial staging and follow-up of anal cancer. The findings of the present study are that PET/CT upstages in 20% of cases and downstages in 27% of cases when used for initial staging, altering the management in 37%. When used for follow-up, it upstages in 11% and downstages in 6% of cases, and alters the management in 17%. The effects of PET on staging anal cancer in the few studies published to date are included in Table 5 for comparison. The results of the present study for upstaging are in keeping with those of the other studies. Where there are significant differences these can be explained by technicalities. For example, in the present study, lymph nodes greater than 10 mm are considered to be pathological on CT, whereas Winton et al. used 15 mm. PET/CT upstaged fewer cases in this study than in Cotter et al., but they compared PET to CT only, whereas the present study compared to CT and MRI making staging by conventional imaging more accurate. In this study, PET/CT was used, as opposed to just PET, allowing attenuation correction. This is now considered the reference standard, but this technology was not available in the majority of the studies above. A previous study by Bannes et al. showed a difference in nodal involvement resulting in a 23% difference in radiotherapy fields between PET/CT and PET or CT separately.8 There are few figures available for comparison of the effect of PET/CT on downstaging anal tumours. This is an important role for PET/CT, and can have an equally significant impact on management. The frequency in which tumours were downstaged as a result of PET/CT in this study is high. Again this is probably due to overcalling nodes on CT due to the 10 mm cut-off used for significance by size criteria. The practice of the present authors is also different from some previous studies, in that follow-up PET/CT is more selectively used, mostly after MDT discussion. That the present results show a higher percentage of cases in which
538
I.T. Wells, B.M. Fox / Clinical Radiology 67 (2012) 535e540
Figure 1 Example of a patient with unsuspected distant nodal metastasis found on PET/CT. PET/CT shows avid FDG uptake (SUV max 8.9) in a tiny aortocaval node (a), which was considered normal on CT performed 4 weeks previously (b) (arrow). There is also avid FDG uptake in a left para-aortic node on PET/CT (c). This node measured 7 mm on CT and was also considered normal (d) (arrow).
PET/CT changed the tumour stage, goes some way to validate this practice. The financial savings made by being more selective in follow-up PET/CT partially offsets the cost of the MDT in person hours. A single PET examination obtained from an independent provider currently costs between
£800 and £1200 depending on the location in the UK and the provider. The breakdown of results by T stage was undertaken because other studies have suggested that PET is more likely to change the overall stage of an anal cancer if it is of higher
Table 5 The effect of positron-emission tomography (PET) on staging anal cancer in the few studies published to date. Authors
Country
Date
No. of patients
Upstaged by PET or PET/CT (%)
Downstaged by PET or PET/CT (%)
Change in management (%)
The present study Nguyen et al.3 Winton et al.2
UK Australia Australia
2008e2011 1996e2006 1997e2005
17 17 15
19 e 8
29 19 16
Trautmann et al.4 Cotter et al.5 Vercellino et al.6
America America France
1999e2002 2003e2005 2004e2008
24 25 e
e e e
e e 20
Iagaru et al.7 Bannas et al.8 Krengli et al.13
America Germany Italy
2009 2010 2005e2008
44 50dcompared to CT only 61dnot all with conventional imaging and PET 21dcompared to CT only 41dcompared to CT only 44 comparison to conventional imaging not made 8 22 27dcompared to CT only
e 9 18.5
e 18 e
38 e 4
Data are overall figures, including initial staging and follow-up examinations.
I.T. Wells, B.M. Fox / Clinical Radiology 67 (2012) 535e540
T stage. A relatively high proportion of T3 tumours (five of 15) were upstaged by PET/CT, but the numbers involved are too small to draw reliable conclusions. Similarly, only 20% of N0 tumours were upstaged by PET/CT, compared to 45 and 47% of N1 and N2 tumours, respectively. Whether distant metastases were present on conventional imaging prior to PET/CT seems not to be useful as a discriminator of whether PET/CT is likely to change the tumour stage, as roughly similar numbers of M1 tumours were downstaged as M0 tumours were upstaged. With recent change in practice towards chemo/radiotherapy and away from surgery as the primary treatment for anal cancer, accurate nodal staging is of increased importance, and this is a key role of PET/CT in anal cancer. PET/CT has little impact on T-staging, as this is purely based on tumour size, which is more accurately determined using MRI than PET/CT. Perirectal nodes are notoriously hard to distinguish from the primary tumour on PET/CT, but this is of little consequence as they will be included in the radiotherapy field anyway. It is important to remember that PET/CT findings are not always clear cut, and should be discussed, preferably in an MDT meeting with a radiologist experienced in interpreting PET imaging. There is a tendency amongst some clinicians to assume that PET/CT is highly specific, but a study by Mistrangelo et al.9 of 27 patients who underwent sentinel node biopsy and PET/CT in anal cancer showed four PET/CT false positives, giving a specificity of 83%. A more recently published study by Vercellino et al.6 compared follow-up PET in 44 anal cancer patients to histology or 6 month follow-up, and calculated a specificity of 81%. Surgical nodal staging and histological confirmation was not performed in the present study so the specificity of PET/CT based on the present results was not calculated. In one case, concern was raised over an inguinal lymph with an SUV of 3.2, but FNA was negative. Several PET/CTs were requested to assess lung nodules identified on staging CT but which were too small for PET/CT to characterize reliably, and interval CT would have been more appropriate. In all the initial staging examinations, the primary tumours were visible on PET/CT, except those that had undergone excision biopsy. This is in keeping with the results of previous studies (93% visible in Vercellino et al.,6 91% in Cotter et al.,5 98% in Nyugen et al.3 and 100% in Winton et al.2). This helps to validate the use of PET/CT in anal cancer, although it has no clinical impact as T-staging is based only on tumour size. Three previous studies have shown that the primary tumour is identified much less frequently by CT (in 58,3 59,5 and 76%10 of cases). There is some evidence (from a small study by Kidd et al.11) that the SUV(max) of the primary tumour is a valuable biomarker of anal cancer prognosis. Of interest is the relatively low incidence of PET/CT changing tumour stage compared to conventional imaging in follow-up cases, despite “selecting” suitable cases by MDT discussion. This may be due to more accurate reporting of CT findings in follow-up studies, perhaps because more previous imaging is available for comparison. For example, previous imaging would allow more certainty
539
about whether a low-attenuation liver lesion is pathological on CT if it was not present previously, or a lung nodule is pathological if it is enlarging d conclusions that could not be drawn on initial staging CT when there may be no previous imaging. If CT reporting is more accurate there is less scope for PET/CT to change it. Although, in all follow-up cases where the stage was altered by PET/CT, it had an effect on management, which was not the case in every initial staging study. It could be argued, therefore, that PET/CT should not be used in the routine follow-up of anal cancer, but that in selected cases it is very useful as a problem-solving tool. The conventional imaging used for staging anal cancer in this study and in many institutions throughout the world includes CT and MRI. In this study, nothing of clinical significance was identified on CT that was not subsequently seen on PET/CT. Where metastases were described on CT they were all also seen on PET/CT or were shown to be false positives by PET/CT (and confirmed by follow-up). All suspicious regional lymph nodes on CT that were not active on PET/CT were treated as uninvolved, so had no impact on management. The poor spatial resolution of the primary tumour on PET/CT is not relevant as this is better evaluated by MRI. The 2010 anal cancer guidelines from the panel of the National Comprehensive Cancer Network stated that the panel “do not consider PET/CT to be a replacement for a diagnostic CT”.12 The present data suggest that it may be reasonable to perform just MRI and PET/CT in the initial staging of anal cancer, and omit the conventional CT examination. This would be even more appropriate if the CT component of PET/CT were to be contrast enhanced. In conclusion, in 2010 the panel of the National Comprehensive Cancer Network issued guidelines on the investigation and management of anal cancer. The panel stated that “the routine use of PET/CT for staging or treatment planning has not been validated”.12 The present study has added to the limited existing data on the role of PET/CT in the initial staging and subsequent follow-up of anal cancer. The findings of the present study are in keeping with the small number of previous studies; PET/CT alters the initial staging of anal cancer in a sufficient number of cases to conclude that it should be used routinely in the initial staging of anal cancer, where available. In particular, three unsuspected distant nodal metastases were found on PET/ CT, significantly altering management. (In three other cases in which unsuspected distant nodal involvement was found, management was not altered.) The suggestion that PET/CT alters the stage more frequently in higher stage tumours cannot be reliably confirmed or refuted as the numbers involved were too small to reliably stratify. Anal cancers are so rare that it is unlikely a study in a single institution will be able to answer this question in a reasonable timeframe, and meta-analysis of the limited published data may be required. The role of PET/CT in follow-up of anal cancer is not as clear as in initial staging as it altered the tumour stage less frequently. It is likely that long-term surveillance with PET/CT will occasionally detect early asymptomatic recurrences that are amenable to curative salvage surgery, but this study suggests that it is very uncommon that these are not also
540
I.T. Wells, B.M. Fox / Clinical Radiology 67 (2012) 535e540
evident on conventional imaging. Where PET/CT is of particular benefit, however, is in further characterizing findings of uncertain significance on MRI, allowing more sensitive detection of recurrence but also avoiding unnecessary biopsy. Routine follow-up with PET/CT may not be justified, therefore, but selected use is of definite benefit in problem solving or if salvage surgery is planned, after MDT discussion.
Acknowledgements The authors thank Dr T. Sulkin of the Royal Cornwall Hospital, Dr S. Higgins of South Devon Hospital, Dr A. Spiers of the Royal Devon and Exeter Hospital, and Dr T. Grunning of Derriford Hospital for their assistance in identifying cases, and Dr D. Sherriff of Derriford Hospital for help with understanding radiotherapy in anal cancer.
References 1. Johnson LG, Madeleine MM, Newcomer LM, et al. Anal cancer incidence and survival: the surveillance, epidemiology, and end results experience, 1973e2000. Cancer 2004;101:281e8. 2. Winton E, Heriot AG, Ng M, et al. The impact of 18-fluorodeoxyglucose positron emission tomography on the staging, management and outcome of anal cancer. Br J Cancer 2009;100:693e700.
3. Nguyen BT, Joon DL, Khoo V, et al. Assessing the impact of FDG-PET in the management of anal cancer. Radiother Oncol 2008;87:376e82. 4. Trautmann TG, Zuger JH. Positron emission tomography for pretreatment staging and posttreatment evaluation in cancer of the anal canal. Mol Imaging Biol 2005;7:309e13. 5. Cotter SE, Grigsby PW, Siegel BA, et al. FDG-PET/CT in the evaluation of anal carcinoma. Int J Radiat Oncol Biol Phys 2006;65:720e5. 6. Vercellino L, Montravers F, de Parades V, et al. Impact of FDG PET/CT in the staging and the follow-up of anal carcinoma. Int J Colorectal Dis 2011;26:201e10. 7. Iagaru A, Kundu R, Jadvar H, et al. Evaluation by 18F-FDG-PET of patients with anal squamous cell carcinoma. Hell J Nucl Med 2009;12:26e9. 8. Bannas P, Weber C, Adam G, et al. Contrast-enhanced [(18)F] fluorodeoxyglucose-positron emission tomography/computed tomography for staging and radiotherapy planning in patients with anal cancer. Int J Radiat Oncol Biol Phys. 2011;81:445e51. M, et al. Comparison of positron emission 9. Mistrangelo M, Pelosi E, Bello tomography scanning and sentinel node biopsy in the detection of inguinal node metastases in patients with anal cancer. Int J Radiat Oncol Biol Phys. 2010;77:73e8. 10. Scherrer A, Reboul F, Martin D, et al. CT of malignant anal canal tumors. RadioGraphics 1990;10:433e53. 11. Kidd EA, Dehdashti F, Siegel BA, et al. Anal cancer maximum F-18 fluorodeoxyglucose uptake on positron emission tomography is correlated with prognosis. Radiother Oncol 2010;95:288e91. 12. Engstrom PF, Arnoletti JP, Benson AB, et al. Anal carcinoma. J Natl Compr Canc Netw 2010;8:106e20. 13. Krengli M, Milia ME, Turri L, et al. FDG-PET/CT imaging for staging and target volume delineation in conformal radiotherapy of anal carcinoma. Radiat Oncol 2010;5:10.
Contents lists available at SciVerse ScienceDirect
Clinical Radiology journal homepage: www.clinicalradiologyonline.net
PET/CT in anal cancer d is it worth doing? I.T. Wells*, B.M. Fox Department of Radiology, Derriford Hospital, Plymouth, Devon, UK
art icl e i nformat ion Article history: Received 29 July 2011 Received in revised form 7 October 2011 Accepted 12 October 2011
AIM: To evaluate the role of 2-[18F]-fluoro-2-deoxy-D-glucose (FDG) positron-emission tomography (PET)/ computed tomography (CT) in the current multidisciplinary management of anal cancer, both in initial staging and in follow-up post-treatment. MATERIALS AND METHODS: All patients referred to the region-wide multidisciplinary meeting for anal cancer during the study period received PET/CT imaging in addition to conventional imaging [CT and magnetic resonance imaging (MRI)]. Whether PET/CT altered the stage of the tumour from that suggested by conventional imaging was retrospectively assessed. The effect on management was evaluated. RESULTS: Fifty PET/CT examinations were performed on 44 patients with anal cancer. Thirty were part of initial staging, and 20 were post-chemo/radiotherapy or surgery. Two PET/CTs produced inadequate contemporaneous conventional imaging to allow comparison. Overall PET/CT increased the stage of the anal cancer in 17% of cases (8/48), decreased the stage in 19% (9/48), and did not alter the stage in 65% (31/48). The tumour stage was altered more frequently in initial staging than in follow up imaging. The PET/CT findings altered patient management in 29% (14/48) of cases. The majority (11) of these were cases in which PET/CT was used as part of initial staging. CONCLUSION: PET/CT alters the initial staging sufficiently frequently that it should be used routinely in anal cancer, where it is available. The role of PET/CT in the follow-up of anal cancer is not as clear. Routine follow-up with PET/CT may not be justified, but selected use is of definite benefit in problem solving or if salvage surgery is planned, after multidisciplinary discussion. Ó 2011 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
Introduction Anal cancer is rare, with an incidence of 1e2 cases per 100,000 in the UK, although this is believed to be increasing with increased exposure to human immunodeficiency virus and human papilloma virus.1 Limited data exist on the best imaging strategy for the initial staging and follow-up of anal cancer. Several small series (the largest by Winton et al. consisting of 61 examinations, in the Peter MacCallum * Guarantor and correspondent: I.T. Wells, Department of Radiology, Derriford Hospital, Plymouth, Devon PL6 8DH, UK. Tel.: þ44 (0) 1752437437; fax: þ44 (0) 1752315300. E-mail address: [email protected] (I.T. Wells).
Cancer Centre Australia)2 have shown that 2-[18F]-fluoro2-deoxy-D-glucose positron-emission tomography (FDG PET) alters the initial staging of anal cancer in a significant number of patients. A few small retrospective series comment on its role in the follow-up of anal cancer patients. In the southwest of England, cases from five sites are discussed at a monthly anal cancer multidisciplinary team (MDT) meeting. This has provided the opportunity to evaluate the role of FDG PET/computed tomography (CT) in current multidisciplinary management of anal cancer, both in initial staging and in follow-up post-treatment. This is particularly relevant at a time when financial resources are stretched, as PET/CT remains an expensive investigation. The experiences in the southwest of England are also widely
0009-9260/$ e see front matter Ó 2011 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.crad.2011.10.030
536
I.T. Wells, B.M. Fox / Clinical Radiology 67 (2012) 535e540
applicable; although patients are discussed at a region-wide MDT meeting, they are imaged and managed at their local hospitals. Through reviewing the MDT discussion process, the present authors have been able to evaluate not only the effect of PET/CT on cancer stage, but on actual patient management. It also allows some reflection on the usefulness of the MDT discussion process, when increasing demands on the time of radiologists, pathologists, oncologists, and surgeons are putting strain on the continued commitment to this model of patient care.
proceed to chemo/radiotherapy or operation, changed the radiotherapy field by detecting unsuspected nodal involvement, or changed a plan to resect liver or lung metastases (where this was previously being considered). Also recorded was the frequency of the primary anal tumour being identified on PET/CT, both on initial staging images and on follow-up images. Extra investigations suggested by the PET/CT images for incidental findings were also documented to allow a qualitative assessment of the financial impact of this.
Results Materials and methods Local review deemed that full ethics approval was not required as the nature of the study is a retrospective audit. All PET/CT examinations for patients with anal cancer performed in the southwest of England since the service started in 2008 were identified by performing a radiology information systems search, and cross-referencing this with records from the anal cancer MDT and the personal records kept by the radiology consultants reporting PET/CT. For each patient identified, the conventional imaging was reviewed and TNM classification and stage of the anal tumour retrospectively recorded. This was done primarily by reading the reports written at the time, with re-review of the images only when the reports were unclear. Conventional imaging consisted of CT (thorax abdomen, and pelvis with intravenous contrast media, and thin-section multiplanar reformat), plus MRI pelvis (small field of view axial, coronal and sagittal T2, large field of view T1 and T2). This region-wide common approach was agreed through the MDT. In all the cases included in the study, conventional imaging was performed within 6 weeks of the PET/CT. If the time gap was larger, or if there was any significant change in therapy between the conventional imaging and the PET/CT being performed, the case was not used. Staging was performed using the sixth edition of the American Joint Committee on Cancer guidelines. The reports of the PET/CTs were then reviewed and TNM classification used to asses tumour status. This was recorded on a separate spreadsheet. Whether the patients were pre or post-initial treatment at the time of the examination was recorded. All PET/CT examinations were performed on a Philips Gemini portable scanner (Philips Medical Systems, Cleveland, OH, USA) after intravenous injection of FDG. CT was obtained without intravenous contrast medium and used for attenuation correction. The frequency with which PET/CT altered the stage of the tumour from the stage suggested by conventional imaging was assessed. These figures were analysed overall and after stratification by initial T stage, N stage and M stage. The subsequent management of the patients was evaluated by looking through subsequent imaging, request cards, notes from the anal cancer MDT meeting, and patient notes where further information was required. The management was considered to be changed if the PET/CT altered a plan to
During the period from June 2008 to May 2011, 50 PET/CT examinations were performed on 44 different patients in the southwest of England with anal cancer. Thirty of these were part of initial staging, and 20 were post-chemo/ radiotherapy or surgery. The follow-up ranged from 6 weeks to 2 years post-treatment. Two PET/CTs produced inadequate contemporaneous conventional imaging to allow comparison. The results are shown in Table 1. Overall PET/CT increased the stage of the anal cancer in 17% of cases (8/48), decreased the stage in 19% (9/48), and did not alter the stage in 65% (31/48). If only the cases in which the PET/CT was performed for initial staging are included, PET/CT increased the overall stage of the anal cancer in 20% of cases (6/30), decreased the stage in 27% (8/30), and did not alter the stage in 53% (16/30). If only the cases in which the PET/CT was performed for follow-up are analysed, PET/CT increased the overall stage in 11% of cases (2/18), decreased the stage in 6% (1/18), and did not alter the stage in 83% (15/18). If the data are broken down by T-stage, the overall stage was changed (either up or down) in 20% (4/20) of T1 tumours, 45% (5/11) of T2 tumours, and 47% (7/15) of T3 tumours. Only one Tx and one T4 tumours were scanned, and the stages of these were unchanged by PET/CT. This is presented in Table 2. If the data are broken down by N-stage, the nodal stage was changed (either up or down) in 12% (3/26) of N0 tumours, 50% (3/6) of N1 tumours, 50% (6/ 12) of N2, and 0% (0/4) of N3 tumours. This is presented in Table 3. Overall, the N stage was altered by PET/CT in 25% (12/48) of cases. If the data are broken down by M-stage, the metastatic stage was upstaged in 18% (6/33) of M0 tumours, and downstaged in 27% (4/15) of M1 tumours. This is presented in Table 4. Overall, the M stage was altered by PET/CT in 21% (10/48) of cases. Table 1 Summary of the effects of positron-emission tomography (PET)/computed tomography (CT) on anal cancer stage compared to conventional imaging.
Overall, including all examinations Only including initial staging examinations Only including follow-up examinations
Upstaged by PET/CT
Downstaged by PET/CT
Unchanged by PET/CT
17% (8/48)
19% (9/48)
65% (31/48)
20% (6/30)
27% (8/30)
53% (16/30)
11% (2/18)
6% (1/18)
83% (15/18)
I.T. Wells, B.M. Fox / Clinical Radiology 67 (2012) 535e540 Table 2 Effect of positron-emission tomography (PET)/computed tomography (CT) on overall stage of anal tumour compared to conventional imaging, broken down by T stage.
Upstaged Downstaged Unchanged
Tx
T1
T2
T3
T4
Total T0e4
0 0 1
1 3 16
1 4 6
5 2 8
0 0 1
7 9 32
The findings of the PET/CT examinations altered patient management in 29% (14/48) of cases. The majority (11) of these were cases in which PET/CT was used as part of initial staging. The remaining three cases in which management was altered by the PET/CT were follow-up imaging after chemo/radiotherapy. Of the 14 cases where management was changed by the PET/CT findings: three were of patients with unsuspected distant nodal metastases found on PET/ CT, in the internal iliac, para aortic and aortocaval regions (Fig 1), and at the aortic bifurcation; one was a regional (inguinal) node not suspected on conventional imaging but with avid uptake on PET/CT; one was a case in which both new regional nodes and distant nodal metastases (paraortic) were found on PET/CT; one was an inguinal node not suspected on conventional imaging but with some metabolic activity on PET/CT (SUV 3.2), for which fine-needle aspiration (FNA) was performed showing this to be a reactive node; three were of lymph-node suspicious on conventional imaging but not metabolically active on PET/CT, which altered the radiotherapy planning field. Subsequent follow-up has confirmed non-involvement in two of these. It is too early to prove this in the other case; one was a case in which a liver metastasis suspected on conventional imaging was negative on PET/CT, but it picked up a new node altering the radiotherapy field (although histological proof of involvement could not be obtained); one was a case in which a liver metastasis though to be single on conventional imaging and planned for resection was found to be multiple on PET/CT (confirmed by MRI); three were of a metastasis suspected on CT but not metabolically active on PET/CT (two in the liver also assessed by MRI, one in the bone). Follow-up of greater than a year was available in two of these three cases to show that these were indeed not metastatic lesions. The primary tumour was seen to be metabolically active on PET/CT in 93% (28/30) of examinations performed as part of the initial staging. In the two initial staging examinations in which the primary was not identified prior excision biopsy had been performed. The primary tumour was evident on 56% (10/18) of the follow-up studies. Table 3 Effect of positron-emission tomography (PET)/computed tomography (CT) on nodal stage of anal tumour compared to conventional imaging, broken down by N stage.
Upstaged Downstaged Unchanged
N0
N1
N2
N3
Total N0e3
3 0 23
1 2 3
2 4 6
0 0 4
6 6 36
537
Table 4 Effect of positron-emission tomography (PET)/computed tomography (CT) on metastatic stage of anal tumour compared to conventional imaging, broken down by M stage.
Upstaged Downstaged Unchanged
M0
M1
Total M0e1
6 0 27
0 4 11
6 4 38
Eight further investigations were suggested in the reports of the PET/CTs: three upper GI endoscopies for uptake in the lower oesophageal sphincter, stomach, and duodenum (two performed with no abnormality identified), three colonoscopies for possible polyps (two performed with a polyp identified on one), one groin node FNA (negative), and one parotid ultrasound (normal).
Discussion This retrospective study adds to the limited data available on the use of PET/CT in the initial staging and follow-up of anal cancer. The findings of the present study are that PET/CT upstages in 20% of cases and downstages in 27% of cases when used for initial staging, altering the management in 37%. When used for follow-up, it upstages in 11% and downstages in 6% of cases, and alters the management in 17%. The effects of PET on staging anal cancer in the few studies published to date are included in Table 5 for comparison. The results of the present study for upstaging are in keeping with those of the other studies. Where there are significant differences these can be explained by technicalities. For example, in the present study, lymph nodes greater than 10 mm are considered to be pathological on CT, whereas Winton et al. used 15 mm. PET/CT upstaged fewer cases in this study than in Cotter et al., but they compared PET to CT only, whereas the present study compared to CT and MRI making staging by conventional imaging more accurate. In this study, PET/CT was used, as opposed to just PET, allowing attenuation correction. This is now considered the reference standard, but this technology was not available in the majority of the studies above. A previous study by Bannes et al. showed a difference in nodal involvement resulting in a 23% difference in radiotherapy fields between PET/CT and PET or CT separately.8 There are few figures available for comparison of the effect of PET/CT on downstaging anal tumours. This is an important role for PET/CT, and can have an equally significant impact on management. The frequency in which tumours were downstaged as a result of PET/CT in this study is high. Again this is probably due to overcalling nodes on CT due to the 10 mm cut-off used for significance by size criteria. The practice of the present authors is also different from some previous studies, in that follow-up PET/CT is more selectively used, mostly after MDT discussion. That the present results show a higher percentage of cases in which
538
I.T. Wells, B.M. Fox / Clinical Radiology 67 (2012) 535e540
Figure 1 Example of a patient with unsuspected distant nodal metastasis found on PET/CT. PET/CT shows avid FDG uptake (SUV max 8.9) in a tiny aortocaval node (a), which was considered normal on CT performed 4 weeks previously (b) (arrow). There is also avid FDG uptake in a left para-aortic node on PET/CT (c). This node measured 7 mm on CT and was also considered normal (d) (arrow).
PET/CT changed the tumour stage, goes some way to validate this practice. The financial savings made by being more selective in follow-up PET/CT partially offsets the cost of the MDT in person hours. A single PET examination obtained from an independent provider currently costs between
£800 and £1200 depending on the location in the UK and the provider. The breakdown of results by T stage was undertaken because other studies have suggested that PET is more likely to change the overall stage of an anal cancer if it is of higher
Table 5 The effect of positron-emission tomography (PET) on staging anal cancer in the few studies published to date. Authors
Country
Date
No. of patients
Upstaged by PET or PET/CT (%)
Downstaged by PET or PET/CT (%)
Change in management (%)
The present study Nguyen et al.3 Winton et al.2
UK Australia Australia
2008e2011 1996e2006 1997e2005
17 17 15
19 e 8
29 19 16
Trautmann et al.4 Cotter et al.5 Vercellino et al.6
America America France
1999e2002 2003e2005 2004e2008
24 25 e
e e e
e e 20
Iagaru et al.7 Bannas et al.8 Krengli et al.13
America Germany Italy
2009 2010 2005e2008
44 50dcompared to CT only 61dnot all with conventional imaging and PET 21dcompared to CT only 41dcompared to CT only 44 comparison to conventional imaging not made 8 22 27dcompared to CT only
e 9 18.5
e 18 e
38 e 4
Data are overall figures, including initial staging and follow-up examinations.
I.T. Wells, B.M. Fox / Clinical Radiology 67 (2012) 535e540
T stage. A relatively high proportion of T3 tumours (five of 15) were upstaged by PET/CT, but the numbers involved are too small to draw reliable conclusions. Similarly, only 20% of N0 tumours were upstaged by PET/CT, compared to 45 and 47% of N1 and N2 tumours, respectively. Whether distant metastases were present on conventional imaging prior to PET/CT seems not to be useful as a discriminator of whether PET/CT is likely to change the tumour stage, as roughly similar numbers of M1 tumours were downstaged as M0 tumours were upstaged. With recent change in practice towards chemo/radiotherapy and away from surgery as the primary treatment for anal cancer, accurate nodal staging is of increased importance, and this is a key role of PET/CT in anal cancer. PET/CT has little impact on T-staging, as this is purely based on tumour size, which is more accurately determined using MRI than PET/CT. Perirectal nodes are notoriously hard to distinguish from the primary tumour on PET/CT, but this is of little consequence as they will be included in the radiotherapy field anyway. It is important to remember that PET/CT findings are not always clear cut, and should be discussed, preferably in an MDT meeting with a radiologist experienced in interpreting PET imaging. There is a tendency amongst some clinicians to assume that PET/CT is highly specific, but a study by Mistrangelo et al.9 of 27 patients who underwent sentinel node biopsy and PET/CT in anal cancer showed four PET/CT false positives, giving a specificity of 83%. A more recently published study by Vercellino et al.6 compared follow-up PET in 44 anal cancer patients to histology or 6 month follow-up, and calculated a specificity of 81%. Surgical nodal staging and histological confirmation was not performed in the present study so the specificity of PET/CT based on the present results was not calculated. In one case, concern was raised over an inguinal lymph with an SUV of 3.2, but FNA was negative. Several PET/CTs were requested to assess lung nodules identified on staging CT but which were too small for PET/CT to characterize reliably, and interval CT would have been more appropriate. In all the initial staging examinations, the primary tumours were visible on PET/CT, except those that had undergone excision biopsy. This is in keeping with the results of previous studies (93% visible in Vercellino et al.,6 91% in Cotter et al.,5 98% in Nyugen et al.3 and 100% in Winton et al.2). This helps to validate the use of PET/CT in anal cancer, although it has no clinical impact as T-staging is based only on tumour size. Three previous studies have shown that the primary tumour is identified much less frequently by CT (in 58,3 59,5 and 76%10 of cases). There is some evidence (from a small study by Kidd et al.11) that the SUV(max) of the primary tumour is a valuable biomarker of anal cancer prognosis. Of interest is the relatively low incidence of PET/CT changing tumour stage compared to conventional imaging in follow-up cases, despite “selecting” suitable cases by MDT discussion. This may be due to more accurate reporting of CT findings in follow-up studies, perhaps because more previous imaging is available for comparison. For example, previous imaging would allow more certainty
539
about whether a low-attenuation liver lesion is pathological on CT if it was not present previously, or a lung nodule is pathological if it is enlarging d conclusions that could not be drawn on initial staging CT when there may be no previous imaging. If CT reporting is more accurate there is less scope for PET/CT to change it. Although, in all follow-up cases where the stage was altered by PET/CT, it had an effect on management, which was not the case in every initial staging study. It could be argued, therefore, that PET/CT should not be used in the routine follow-up of anal cancer, but that in selected cases it is very useful as a problem-solving tool. The conventional imaging used for staging anal cancer in this study and in many institutions throughout the world includes CT and MRI. In this study, nothing of clinical significance was identified on CT that was not subsequently seen on PET/CT. Where metastases were described on CT they were all also seen on PET/CT or were shown to be false positives by PET/CT (and confirmed by follow-up). All suspicious regional lymph nodes on CT that were not active on PET/CT were treated as uninvolved, so had no impact on management. The poor spatial resolution of the primary tumour on PET/CT is not relevant as this is better evaluated by MRI. The 2010 anal cancer guidelines from the panel of the National Comprehensive Cancer Network stated that the panel “do not consider PET/CT to be a replacement for a diagnostic CT”.12 The present data suggest that it may be reasonable to perform just MRI and PET/CT in the initial staging of anal cancer, and omit the conventional CT examination. This would be even more appropriate if the CT component of PET/CT were to be contrast enhanced. In conclusion, in 2010 the panel of the National Comprehensive Cancer Network issued guidelines on the investigation and management of anal cancer. The panel stated that “the routine use of PET/CT for staging or treatment planning has not been validated”.12 The present study has added to the limited existing data on the role of PET/CT in the initial staging and subsequent follow-up of anal cancer. The findings of the present study are in keeping with the small number of previous studies; PET/CT alters the initial staging of anal cancer in a sufficient number of cases to conclude that it should be used routinely in the initial staging of anal cancer, where available. In particular, three unsuspected distant nodal metastases were found on PET/ CT, significantly altering management. (In three other cases in which unsuspected distant nodal involvement was found, management was not altered.) The suggestion that PET/CT alters the stage more frequently in higher stage tumours cannot be reliably confirmed or refuted as the numbers involved were too small to reliably stratify. Anal cancers are so rare that it is unlikely a study in a single institution will be able to answer this question in a reasonable timeframe, and meta-analysis of the limited published data may be required. The role of PET/CT in follow-up of anal cancer is not as clear as in initial staging as it altered the tumour stage less frequently. It is likely that long-term surveillance with PET/CT will occasionally detect early asymptomatic recurrences that are amenable to curative salvage surgery, but this study suggests that it is very uncommon that these are not also
540
I.T. Wells, B.M. Fox / Clinical Radiology 67 (2012) 535e540
evident on conventional imaging. Where PET/CT is of particular benefit, however, is in further characterizing findings of uncertain significance on MRI, allowing more sensitive detection of recurrence but also avoiding unnecessary biopsy. Routine follow-up with PET/CT may not be justified, therefore, but selected use is of definite benefit in problem solving or if salvage surgery is planned, after MDT discussion.
Acknowledgements The authors thank Dr T. Sulkin of the Royal Cornwall Hospital, Dr S. Higgins of South Devon Hospital, Dr A. Spiers of the Royal Devon and Exeter Hospital, and Dr T. Grunning of Derriford Hospital for their assistance in identifying cases, and Dr D. Sherriff of Derriford Hospital for help with understanding radiotherapy in anal cancer.
References 1. Johnson LG, Madeleine MM, Newcomer LM, et al. Anal cancer incidence and survival: the surveillance, epidemiology, and end results experience, 1973e2000. Cancer 2004;101:281e8. 2. Winton E, Heriot AG, Ng M, et al. The impact of 18-fluorodeoxyglucose positron emission tomography on the staging, management and outcome of anal cancer. Br J Cancer 2009;100:693e700.
3. Nguyen BT, Joon DL, Khoo V, et al. Assessing the impact of FDG-PET in the management of anal cancer. Radiother Oncol 2008;87:376e82. 4. Trautmann TG, Zuger JH. Positron emission tomography for pretreatment staging and posttreatment evaluation in cancer of the anal canal. Mol Imaging Biol 2005;7:309e13. 5. Cotter SE, Grigsby PW, Siegel BA, et al. FDG-PET/CT in the evaluation of anal carcinoma. Int J Radiat Oncol Biol Phys 2006;65:720e5. 6. Vercellino L, Montravers F, de Parades V, et al. Impact of FDG PET/CT in the staging and the follow-up of anal carcinoma. Int J Colorectal Dis 2011;26:201e10. 7. Iagaru A, Kundu R, Jadvar H, et al. Evaluation by 18F-FDG-PET of patients with anal squamous cell carcinoma. Hell J Nucl Med 2009;12:26e9. 8. Bannas P, Weber C, Adam G, et al. Contrast-enhanced [(18)F] fluorodeoxyglucose-positron emission tomography/computed tomography for staging and radiotherapy planning in patients with anal cancer. Int J Radiat Oncol Biol Phys. 2011;81:445e51. M, et al. Comparison of positron emission 9. Mistrangelo M, Pelosi E, Bello tomography scanning and sentinel node biopsy in the detection of inguinal node metastases in patients with anal cancer. Int J Radiat Oncol Biol Phys. 2010;77:73e8. 10. Scherrer A, Reboul F, Martin D, et al. CT of malignant anal canal tumors. RadioGraphics 1990;10:433e53. 11. Kidd EA, Dehdashti F, Siegel BA, et al. Anal cancer maximum F-18 fluorodeoxyglucose uptake on positron emission tomography is correlated with prognosis. Radiother Oncol 2010;95:288e91. 12. Engstrom PF, Arnoletti JP, Benson AB, et al. Anal carcinoma. J Natl Compr Canc Netw 2010;8:106e20. 13. Krengli M, Milia ME, Turri L, et al. FDG-PET/CT imaging for staging and target volume delineation in conformal radiotherapy of anal carcinoma. Radiat Oncol 2010;5:10.