- Email: [email protected]
Occult Malignancy in Patients With Suspected Paraneoplastic Neurologic Syndromes: Value of Positron Emission Tomography in Diagnosis
ORIGINAL ARTICLE
OCCULT MALIGNANCY IN SUSPECTED PNS
Occult Malignancy in Patients With Suspected Paraneoplastic Neurologic Syndromes: Value of Positron Emission Tomography in Diagnosis RAJESH R. PATEL, MBBS; RATHAN M. SUBRAMANIAM, MBBS, MD; JAYWANT N. MANDREKAR, PHD; JULIE E. HAMMACK, MD; VAL J. LOWE, MD; AND JAMES R. JETT, MD OBJECTIVE: To determine the value of positron emission tomography (PET) in diagnosing occult malignancies in patients with paraneoplastic neurologic syndromes (PNSs) at Mayo Clinic’s site in Rochester, MN. PATIENTS AND METHODS: We retrospectively reviewed the medical charts of all 107 patients who underwent PET from January 1, 2000, to July 31, 2006, for the indication of suspected PNS. Three patients did not meet inclusion criteria. PET results were considered positive if increased fludeoxyglucose F 18 uptake indicated malignancy (24 patients). Results from computed tomography were interpreted as positive if any suspect lesion was consistent with malignancy (26 patients). RESULTS: One hundred four patients with PNS were identified from the PET central database; 73 patients had at least 1 positive result for paraneoplastic antibody, and 31 had antibody-negative PNS. Malignancy was confirmed pathologically in 10 patients, of whom 8 had positive PET results. There were 2 cases of confirmed malignancy (fallopian tube adenocarcinoma and spindle cell uterine carcinoma) for which PET results were negative. Two patients with positive PET results declined biopsy. Computed tomography was able to identify 3 of the 10 malignancies detected. Five cases of malignancy were detected only by PET. All patients with confirmed malignancy had positive results for at least 1 paraneoplastic antibody. One patient with positive results for PNS antibody and negative PET results was diagnosed as having small cell carcinoma on a follow-up PET scan after 27 months. PET had sensitivity, specificity, positive predictive value, and negative predictive value of 80%, 67%, 53%, and 88%, respectively. CONCLUSION: PET scan was shown to be more sensitive than computed tomography for detecting occult malignancy (confirmed by positive test results for autoantibody) among patients with suspected PNS. The greatest clinical utility of PET could be in its high negative predictive value.
Mayo Clin Proc. 2008;83(8):917-922
CT = computed tomography; NPV = negative predictive value; PET = positron emission tomography; PNS = paraneoplastic neurologic syndrome; PPV = positive predictive value
From the Division of Pulmonary and Critical Care Medicine (R.R.P., J.R.J.), Department of Radiology (R.M.S., V.J.L.), Division of Biostatistics (J.N.M.), and Department of Neurology (J.E.H.), Mayo Clinic, Rochester, MN. These data were presented as a poster at the 12th World Conference on Lung Cancer; Seoul, South Korea; September 2-6, 2007. Individual reprints of this article are not available. Address correspondence to James R. Jett, MD, Division of Pulmonary and Critical Care Medicine, Mayo Clinic, 200 First St SW, Rochester, MN 55905 ([email protected]). © 2008 Mayo Foundation for Medical Education and Research
Mayo Clin Proc.
•
P
araneoplastic neurologic syndrome (PNS) refers to signs and symptoms resulting from damage to any part of the nervous system that is remote from the site of malignancy or metastases.1 Its effects on various components of the central and peripheral nervous system can vary substantially.2,3 Tumors are thought to express certain antigens ectopically that have homologues expressed exclusively in the nervous system. Some of these “onco-neural” antigens are normally present in immunologically privileged sites like brain and normal testes. When the immune system encounters these antigens on tumor cells, they are recognized as foreign antigens. Therefore, an immune response is mounted against them.1,2,4 This response leads to generation of various antibodies that can be detected in the serum and cerebrospinal fluid of patients.5,6 These antineuronal antibodies2 can cross the blood-brain barrier and attack the central or peripheral nervous system, producing PNS.7-9 These antibodies are found in many patients with cancer who do not have PNS, and only 0.01% of patients with cancer develop PNS.3 The most common cancers associated with PNS are small cell lung cancer (3%), thymoma (15%), and POEMS syndrome (100%) (polyneuropathy, organomegaly, endocrinopathy, M protein, and skin changes). Frequently, PNS develops before a cancer has been identified. Patients present with the signs and symptoms of PNS that sometimes lead to diagnosis of a malignancy. A range of antineuronal antibodies can be detected in patients with PNS, but they are neither completely sensitive nor specific for a certain type of malignancy10-12; hence, physicians need to search for occult malignancy in these patients.4 Moreover, not all patients with circulating antineuronal antibodies will be found to have malignancy, even after lengthy follow-up.1,12,13 Routine screening tests, such as conventional computed tomography (CT) and magnetic resonance imaging, are often unrewarding in detecting occult malignancies and differentiating benign from malignant lesions. Positron emission tomography (PET) using fludeoxyglucose F 18 has been used to detect occult cancers in some patients.14 Although the European Federation of Neurological Socie-
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
917
OCCULT MALIGNANCY IN SUSPECTED PNS
TABLE 1. Baseline Characteristics of All Patients (N=104)a Age (y), median (range) Male, No. (%) Antibodies,b frequency (%) AChR Binding Ganglionic N ANNA type 1 2 3 Unclassified antineuronal PCA type 1 2 Tr Amphiphysin Collapsin response-mediator protein-5 IgG Striated muscle N-type calcium channel P/Q-type calcium channel Potassium channel Glutamic acid decarboxylase 65 PNS follow-up duration (d), median (range) With malignancy Without malignancy
56 (17-84) 46 (44) 9 (8.7) 5 (4.8) 8 (7.7) 1 (1.0) 0 (0.0) 10 (9.6) 1 (1.0) 1 (1.0) 0 (0.0) 1 (1.0) 11 (10.6) 6 (5.8) 9 (8.7) 9 (8.7) 12 (11.5) 9 (8.7) 770 (22-2069) 337 (22-2519)
RADIOLOGY REVIEW PET results were considered positive if increased fludeoxyglucose F 18 uptake indicated malignancy. Computed tomography was considered positive if any lesion or mass was consistent with malignancy. The CT and PET scans were read initially as a standard clinical test by various readers and then reviewed by a coauthor (R.M.S.) who was blinded to the clinical data. Results from subsequent CT and PET scans for each patient were also recorded. INCLUSION CRITERIA Paraneoplastic neurologic syndrome was defined as the presence of signs and symptoms caused by damage to any part of the nervous system not due to a primary tumor, metastasis, or any other brain, spinal cord, or nerve disorder.1,15,16 The clinical diagnosis of suspected PNS was extracted from the neurologic consultation records. Patients with a final clinical diagnosis of PNS were included in the analysis.
a
AChR = acetylcholine receptor; ANNA = antineuronal antibody; PCA = Purkinje-cell cytoplasmic autoantibody; PNS = paraneoplastic neurologic syndrome. b Some patients had multiple antibodies.
ties recommends the use of PET in evaluating PNS patients for occult malignancy, the sensitivity and specificity of the technique has not been well defined. We conducted this retrospective study to define the clinical utility of PET for detecting occult malignancies in patients with PNS.
STATISTICAL ANALYSES In addition to basic summary statistics, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for detection of malignancy by PET and by CT were estimated along with 95% exact binomial confidence interval. Detection of malignancy by PET and CT was compared using the McNemar test, a test of paired proportions. RESULTS
PATIENTS AND METHODS After approval by the Mayo Clinic Institutional Review Board, the nuclear medicine electronic database at Mayo Clinic’s site in Rochester, MN, was queried for PNS as an indication for PET. A total of 107 patients were identified in the database from January 1, 2000, to July 31, 2006. CHART REVIEW We retrospectively reviewed the medical charts for 107 patients; 3 were ineligible. Data collected for every patient included demographics, date of paraneoplastic antibody test, type of paraneoplastic antibody identified, pathology report of the lesions (if available), and duration of followup for each patient. The paraneoplastic antibody tested and normal titer values are in the Appendix. Presence of at least 1 paraneoplastic antibody with titers exceeding the normal limits was considered a positive result. The duration of follow-up for each patient was calculated from the date of first diagnosis of PNS to the last date of follow-up or death. 918
Mayo Clin Proc.
•
Of the 107 patients identified from the PET database, 104 met the inclusion criteria.15 Three patients were excluded because 1 had metastasis to the brain and 2 did not have a final diagnosis of PNS. Baseline characteristics of all the patients are described in Table 1. Seventy-three patients had a positive result for at least 1 paraneoplastic antibody, and 31 had a clinical diagnosis without a detectable paraneoplastic antibody. PET results were positive in 24 patients, 17 of whom had a positive result for at least 1 paraneoplastic antibody. Concomitant CT data were available for 71 patients. Results from CT were positive in 26 patients. A total of 31 patients underwent biopsy or surgery, and the rest were monitored (Figure 1). Ten patients had histologic or surgical proof of malignancy. Characteristics of all 10 patients are listed in Table 2. PET was able to detect 8 of 10 malignancies diagnosed (Figure 2). Computed tomography, by comparison, detected 3 of 10 malignancies (Figure 2). The 2 false-negative PET results were a 5-mm adenocarcinoma of the fallopian tube (Figure 3) and metastatic retroperitoneal lymph nodes
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
OCCULT MALIGNANCY IN SUSPECTED PNS
107 Patients underwent PET for suspected PNS
3 Did not meet inclusion criteria
104 Met inclusion criteria
73 PNS antibody positive
31 PNS antibody negative
54 PET negative
17 PET positive
7 PET positive
22 PET negative
1 Biopsy or surgery
0 Biopsy or surgery
2 Refused biopsy or surgery 15 Biopsy or surgery
15 Biopsy or surgery
FIGURE 1. Schematic chart of patient distribution. PET = positron emission tomography; PNS = paraneoplastic neurologic syndrome.
TABLE 2. Ten Patients With PNS Diagnosed With Malignancya Age (y)/ sex 36/M
Morvan chorea
83/F
PNS antibody
PET result
CT result
Final diagnosis
Follow-up (d)
Outcome
Positive
Positive
Thymoma
1227
Alive
Chorea
GAD-65 and potassium-channel antibody CRMP-5
Positive
Positive
1299
Dead
68/F
Chorea
ANNA-1
Positive
Negative
2069
Alive
74/F
Cerebellar degeneration Ganglioneuropathy
PCA-1
Negative
Negative
1037
Dead
ANNA-1 and N-type calcium channel ANNA-1
Positive
Negative
Adenocarcinoma of right lower lobe; small cell carcinoma of mediastinal nodes Mixed small and large cell carcinoma Serous carcinoma of right fallopian tube Small cell carcinoma
22
Dead
Positive
Positive
Small cell carcinoma
235
Positive
Negative
Small cell carcinoma
56
Positive Positive
Negative Negative
Negative
Negative
Small cell carcinoma Adenocarcinoma of unknown origin Metastatic spindle cell carcinoma diagnosed at autopsy
64/M 46/F 66/F 64/F 51/F 45/F
a
PNS
Peripheral polyradiculopathy Large-fiber sensory neuropathy Sensory myelopathy Multiple cranial nerve deficit Progressive multifocal neurological deficit
CRMP-5, PCA-2, and antiamphiphysin CRMP-5 ANNA-1 ANNA-1 and potassium-channel antibody
Lost to follow-up Dead
790 751
Alive Alive
50
Dead
ANNA = antineuronal antibody; CRMP = collapsin response-mediator protein; CT = computed tomography; GAD = glutamic acid decarboxylase; PCA = Purkinje-cell cytoplasmic autoantibody; PET = positron emission tomography; PNS = paraneoplastic neurologic syndrome. Mayo Clin Proc.
•
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
919
OCCULT MALIGNANCY IN SUSPECTED PNS
12
Malignancies diagnosed
10 8
6
4 2
0
Total malignancies
PET results positive
CT results positive
FIGURE 2. Distribution of malignancies and positive results from positron emission tomography (PET) and computed tomography (CT).
FIGURE 3. A 74-year-old woman with paraneoplastic neurologic syndrome. Computed tomography and fludeoxyglucose F 18 used with positron emission tomography demonstrated no hypermetabolic focus indicating malignancy. Patient had serous carcinoma of the right fallopian tube.
in a patient with previously resected spindle cell carcinoma of the uterus. Figure 4 represents positive PET results in a patient with an initial negative CT scan. Two patients with
positive results from PET deferred biopsy for unknown reasons and were lost to follow-up. Patients were followed up in a neurology clinic for assessment and treatment of
FIGURE 4. A 64-year-old woman with paraneoplastic neurologic syndrome. Computed tomography and fludeoxyglucose F 18 used with positron emission tomography demonstrated hypermetabolic mediastinal lymphadenopathy (arrows), which was proven to be small cell carcinoma via biopsy.
920
Mayo Clin Proc.
•
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
OCCULT MALIGNANCY IN SUSPECTED PNS
TABLE 3. Estimates of Sensitivity, Specificity, PPV, and NPV of PET and CTa
Sensitivity Specificity PPV NPV a
PET % (95% CI)
CT % (95% CI)
80 (44-98) 67 (43-85) 53 (27-79) 88 (62-99)
30 (7-65) 71 (42-89) 33 (8-70) 68 (45-86)
CI = confidence interval; CT = computed tomography; NPV = negative predictive value; PET = positron emission tomography; PPV = positive predictive value.
their neurologic disease. The median length of follow-up was 337 days (range, 22-2519 days). During the follow-up period, one patient with initial negative results from PET and an antibody result positive for PNS was diagnosed with small cell carcinoma of the lung on a follow-up PET study after 27 months. There were 16 false-positive results from PET. The specificity of PET was 67%, and the PPV was 53%. The sensitivity of PET for malignancy was 80%, and the NPV was 88% (Table 3). The sensitivity, specificity, PPV, and NPV of CT were 30%, 71%, 33%, and 68%, respectively. PET was significantly better than CT for detecting occult malignancy in our patient population (P=.02). DISCUSSION Patients with suspected PNS present a diagnostic dilemma because of the difficulty in detecting a suspected occult malignancy with conventional testing. Although PET has been recommended and used to detect occult malignancies in this patient population, its diagnostic utility has not been well defined.12 We have scrutinized the clinical utility of PET in patients with suspected PNS. Data suggest that PET has a better NPV (88%) and therefore is a better diagnostic tool to rule out occult malignancy, especially of the lung and mediastinum (PET did not miss any of these malignancies). All patients with cancer had a positive test for PNS antibody, and no patients with antibody-negative results were found to have malignancy. If we consider only antibody-positive patients, the diagnostic utility of PET scan is further enhanced. Data also revealed a high false-positive rate of PET, necessitating confirmation with a tissue diagnosis. The clinical dilemma remains for patients with positive antibody results and no cancer detected with PET.17,18 In our study, 62 of 73 antibody-positive PNS patients had no malignancy detected after extensive work-up and follow-up. There is a remote possibility that these patients might have early malignancy, and development of autoantibodies might have resulted in resolution of the tumor Mayo Clin Proc.
•
or rendered it indictable because of its size.13 The Paraneoplastic Neurological Syndrome Euronetwork recommends a follow-up period of 5 years to declare that a patient does not have any occult malignancies.14 We observed only one malignancy in these patients during follow-up. Thus, it can be inferred that not all patients with a positive test for paraneoplastic antibody and PNS will have occult malignancies. Our study is limited by its retrospective nature, the small patient population, and the rarity of this condition. Nevertheless, it remains one of the largest evaluations of the utility of PET for detecting occult malignancy. The high rate of false-positive results from PET and the inability of PET to detect smaller (8-10 mm) tumors limits the test’s value. Some patients in our study had limited follow-up at our institution because longdistance travel and neurologic impairment made return visits difficult. CONCLUSION PET is substantially more sensitive than CT for detecting occult malignancy in patients with suspected PNS. Occult malignancy was detected only in patients with PNS who had a positive test for autoantibody. Patients with suspected PNS and no detectable malignancy should be followed up every 6 to 12 months for a period of at least 5 years as recommended by the European Federation of Neurological Societies. The utility of follow-up CT or PET in these patients could not be determined because of limited data.
REFERENCES 1. Darnell RB, Posner JB. Paraneoplastic syndromes involving the nervous system. N Engl J Med. 2003;349(16):1543-1554. 2. Darnell RB. Onconeural antigens and the paraneoplastic neurologic disorders: at the intersection of cancer, immunity, and the brain. Proc Natl Acad Sci U S A. 1996;93(10):4529-4536. 3. Pittock SJ, Kryzer TJ, Lennon VA. Paraneoplastic antibodies coexist and predict cancer, not neurological syndrome. Ann Neurol. 2004;56(5):715-719. 4. Bredholt G, Storstein A, Haugen M, et al. Detection of autoantibodies to the BTB-kelch protein KLHL7 in cancer sera. Scand J Immunol. 2006;64(3):325-335. 5. Anderson NE, Rosenblum MK, Graus F, Wiley RG, Posner JB. Autoantibodies in paraneoplastic syndromes associated with small-cell lung cancer. Neurology. 1988;38(9):1391-1398. 6. Moll JW, Antoine JC, Brashear HR, et al. Guidelines on the detection of paraneoplastic anti-neuronal-specific antibodies: report from the Workshop to the Fourth Meeting of the International Society of Neuro-Immunology on paraneoplastic neurological disease, held October 22-23, 1994, in Rotterdam, The Netherlands. Neurology. 1995;45(10):1937-1941. 7. Darnell RB. Paraneoplastic neurologic disorders: windows into neuronal function and tumor immunity. Arch Neurol. 2004;61(1):30-32. 8. Albert ML, Darnell RB. Paraneoplastic neurological degenerations: keys to tumour immunity. Nat Rev Cancer. 2004;4(1):36-44.
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
921
OCCULT MALIGNANCY IN SUSPECTED PNS
9. Nardelli E, Curro-Dossi B, Rizzuto N, Graus F. Autoimmune paraneoplastic cerebellar degeneration: immunohistological localization of antibody-binding sites. Clin Neurol Neurosurg. 1995;97(1):89-94. 10. Graus F, Lang B, Pozo-Rosich P, Saiz A, Casamitjana R, Vincent A. P/Q type calcium-channel antibodies in paraneoplastic cerebellar degeneration with lung cancer. Neurology. 2002;59(5):764-766. 11. Rojas-Marcos I, Rousseau A, Keime-Guibert F, et al. Spectrum of paraneoplastic neurologic disorders in women with breast and gynecologic cancer. Medicine (Baltimore). 2003;82(3):216-223. 12. Voltz R, Graus F. Diagnosis and treatment of paraneoplastic neurological disorders. Onkologie. 2004;27(3):253-258. 13. Graus F, Keime-Guibert F, Rene R, et al. Anti-Hu-associated paraneoplastic encephalomyelitis: analysis of 200 patients. Brain. 2001;124(pt 6): 1138-1148. 14. Vedeler CA, Antoine JC, Giometto B, et al, Paraneoplastic Neurological Syndrome Euronetwork. Management of paraneoplastic neurological syndromes: report of an EFNS Task Force. Eur J Neurol. 2006;13(7):682-690. 15. Graus F, Delattre JY, Antoine JC, et al. Recommended diagnostic criteria for paraneoplastic neurological syndromes. J Neurol Neurosurg Psychiatry. 2004;75(8):1135-1140. 16. Bataller L, Graus F, Saiz A, Vilchez JJ, Spanish Opsoclonus-Myoclonus Study Group. Clinical outcome in adult onset idiopathic or paraneoplastic opsoclonus-myoclonus. Brain. 2001;124(pt 2):437-443. 17. Graus F, Bonaventura I, Uchuya M, et al. Indolent anti-Hu-associated paraneoplastic sensory neuropathy. Neurology. 1994;44(12):2258-2261. 18. Darnell RB, DeAngelis LM. Regression of small-cell lung carcinoma in patients with paraneoplastic neuronal antibodies. Lancet. 1993;341(8836):2122.
922
Mayo Clin Proc.
•
APPENDIX. Paraneoplastic Antibodies and Their Normal Valuesa Normal values at Mayo Clinic laboratories
Paraneoplastic antibodies AChR Binding Ganglionic N ANNA type 1 (anti-Hu) 2 (anti-Ri) 3 Unclassified antineuronal
)0.02 nmol/L )0.02 nmol/L <1:240 titer <1:240 titer <1:240 titer Nonspecific IgG in titers higher than 1:240
PCA type 1(anti-Yo) 2 Tr Amphiphysin Collapsin response-mediator receptor-5 (anti-CV-2) Striated muscle N-type calcium channel P/Q-type calcium channel Potassium channel Glutamic acid decarboxylase 65 a
<1:240 titer <1:240 titer <1:240 titer <1:240 titer <1:240 titer <1:60 titer )0.02 nmol/L )0.02 nmol/L Absent <0.02 nmol/L
AChR = acetylcholine receptor; ANNA = antineuronal antibody; PCA = Purkinje-cell cytoplasmic autoantibody
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
OCCULT MALIGNANCY IN SUSPECTED PNS
Occult Malignancy in Patients With Suspected Paraneoplastic Neurologic Syndromes: Value of Positron Emission Tomography in Diagnosis RAJESH R. PATEL, MBBS; RATHAN M. SUBRAMANIAM, MBBS, MD; JAYWANT N. MANDREKAR, PHD; JULIE E. HAMMACK, MD; VAL J. LOWE, MD; AND JAMES R. JETT, MD OBJECTIVE: To determine the value of positron emission tomography (PET) in diagnosing occult malignancies in patients with paraneoplastic neurologic syndromes (PNSs) at Mayo Clinic’s site in Rochester, MN. PATIENTS AND METHODS: We retrospectively reviewed the medical charts of all 107 patients who underwent PET from January 1, 2000, to July 31, 2006, for the indication of suspected PNS. Three patients did not meet inclusion criteria. PET results were considered positive if increased fludeoxyglucose F 18 uptake indicated malignancy (24 patients). Results from computed tomography were interpreted as positive if any suspect lesion was consistent with malignancy (26 patients). RESULTS: One hundred four patients with PNS were identified from the PET central database; 73 patients had at least 1 positive result for paraneoplastic antibody, and 31 had antibody-negative PNS. Malignancy was confirmed pathologically in 10 patients, of whom 8 had positive PET results. There were 2 cases of confirmed malignancy (fallopian tube adenocarcinoma and spindle cell uterine carcinoma) for which PET results were negative. Two patients with positive PET results declined biopsy. Computed tomography was able to identify 3 of the 10 malignancies detected. Five cases of malignancy were detected only by PET. All patients with confirmed malignancy had positive results for at least 1 paraneoplastic antibody. One patient with positive results for PNS antibody and negative PET results was diagnosed as having small cell carcinoma on a follow-up PET scan after 27 months. PET had sensitivity, specificity, positive predictive value, and negative predictive value of 80%, 67%, 53%, and 88%, respectively. CONCLUSION: PET scan was shown to be more sensitive than computed tomography for detecting occult malignancy (confirmed by positive test results for autoantibody) among patients with suspected PNS. The greatest clinical utility of PET could be in its high negative predictive value.
Mayo Clin Proc. 2008;83(8):917-922
CT = computed tomography; NPV = negative predictive value; PET = positron emission tomography; PNS = paraneoplastic neurologic syndrome; PPV = positive predictive value
From the Division of Pulmonary and Critical Care Medicine (R.R.P., J.R.J.), Department of Radiology (R.M.S., V.J.L.), Division of Biostatistics (J.N.M.), and Department of Neurology (J.E.H.), Mayo Clinic, Rochester, MN. These data were presented as a poster at the 12th World Conference on Lung Cancer; Seoul, South Korea; September 2-6, 2007. Individual reprints of this article are not available. Address correspondence to James R. Jett, MD, Division of Pulmonary and Critical Care Medicine, Mayo Clinic, 200 First St SW, Rochester, MN 55905 ([email protected]). © 2008 Mayo Foundation for Medical Education and Research
Mayo Clin Proc.
•
P
araneoplastic neurologic syndrome (PNS) refers to signs and symptoms resulting from damage to any part of the nervous system that is remote from the site of malignancy or metastases.1 Its effects on various components of the central and peripheral nervous system can vary substantially.2,3 Tumors are thought to express certain antigens ectopically that have homologues expressed exclusively in the nervous system. Some of these “onco-neural” antigens are normally present in immunologically privileged sites like brain and normal testes. When the immune system encounters these antigens on tumor cells, they are recognized as foreign antigens. Therefore, an immune response is mounted against them.1,2,4 This response leads to generation of various antibodies that can be detected in the serum and cerebrospinal fluid of patients.5,6 These antineuronal antibodies2 can cross the blood-brain barrier and attack the central or peripheral nervous system, producing PNS.7-9 These antibodies are found in many patients with cancer who do not have PNS, and only 0.01% of patients with cancer develop PNS.3 The most common cancers associated with PNS are small cell lung cancer (3%), thymoma (15%), and POEMS syndrome (100%) (polyneuropathy, organomegaly, endocrinopathy, M protein, and skin changes). Frequently, PNS develops before a cancer has been identified. Patients present with the signs and symptoms of PNS that sometimes lead to diagnosis of a malignancy. A range of antineuronal antibodies can be detected in patients with PNS, but they are neither completely sensitive nor specific for a certain type of malignancy10-12; hence, physicians need to search for occult malignancy in these patients.4 Moreover, not all patients with circulating antineuronal antibodies will be found to have malignancy, even after lengthy follow-up.1,12,13 Routine screening tests, such as conventional computed tomography (CT) and magnetic resonance imaging, are often unrewarding in detecting occult malignancies and differentiating benign from malignant lesions. Positron emission tomography (PET) using fludeoxyglucose F 18 has been used to detect occult cancers in some patients.14 Although the European Federation of Neurological Socie-
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
917
OCCULT MALIGNANCY IN SUSPECTED PNS
TABLE 1. Baseline Characteristics of All Patients (N=104)a Age (y), median (range) Male, No. (%) Antibodies,b frequency (%) AChR Binding Ganglionic N ANNA type 1 2 3 Unclassified antineuronal PCA type 1 2 Tr Amphiphysin Collapsin response-mediator protein-5 IgG Striated muscle N-type calcium channel P/Q-type calcium channel Potassium channel Glutamic acid decarboxylase 65 PNS follow-up duration (d), median (range) With malignancy Without malignancy
56 (17-84) 46 (44) 9 (8.7) 5 (4.8) 8 (7.7) 1 (1.0) 0 (0.0) 10 (9.6) 1 (1.0) 1 (1.0) 0 (0.0) 1 (1.0) 11 (10.6) 6 (5.8) 9 (8.7) 9 (8.7) 12 (11.5) 9 (8.7) 770 (22-2069) 337 (22-2519)
RADIOLOGY REVIEW PET results were considered positive if increased fludeoxyglucose F 18 uptake indicated malignancy. Computed tomography was considered positive if any lesion or mass was consistent with malignancy. The CT and PET scans were read initially as a standard clinical test by various readers and then reviewed by a coauthor (R.M.S.) who was blinded to the clinical data. Results from subsequent CT and PET scans for each patient were also recorded. INCLUSION CRITERIA Paraneoplastic neurologic syndrome was defined as the presence of signs and symptoms caused by damage to any part of the nervous system not due to a primary tumor, metastasis, or any other brain, spinal cord, or nerve disorder.1,15,16 The clinical diagnosis of suspected PNS was extracted from the neurologic consultation records. Patients with a final clinical diagnosis of PNS were included in the analysis.
a
AChR = acetylcholine receptor; ANNA = antineuronal antibody; PCA = Purkinje-cell cytoplasmic autoantibody; PNS = paraneoplastic neurologic syndrome. b Some patients had multiple antibodies.
ties recommends the use of PET in evaluating PNS patients for occult malignancy, the sensitivity and specificity of the technique has not been well defined. We conducted this retrospective study to define the clinical utility of PET for detecting occult malignancies in patients with PNS.
STATISTICAL ANALYSES In addition to basic summary statistics, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for detection of malignancy by PET and by CT were estimated along with 95% exact binomial confidence interval. Detection of malignancy by PET and CT was compared using the McNemar test, a test of paired proportions. RESULTS
PATIENTS AND METHODS After approval by the Mayo Clinic Institutional Review Board, the nuclear medicine electronic database at Mayo Clinic’s site in Rochester, MN, was queried for PNS as an indication for PET. A total of 107 patients were identified in the database from January 1, 2000, to July 31, 2006. CHART REVIEW We retrospectively reviewed the medical charts for 107 patients; 3 were ineligible. Data collected for every patient included demographics, date of paraneoplastic antibody test, type of paraneoplastic antibody identified, pathology report of the lesions (if available), and duration of followup for each patient. The paraneoplastic antibody tested and normal titer values are in the Appendix. Presence of at least 1 paraneoplastic antibody with titers exceeding the normal limits was considered a positive result. The duration of follow-up for each patient was calculated from the date of first diagnosis of PNS to the last date of follow-up or death. 918
Mayo Clin Proc.
•
Of the 107 patients identified from the PET database, 104 met the inclusion criteria.15 Three patients were excluded because 1 had metastasis to the brain and 2 did not have a final diagnosis of PNS. Baseline characteristics of all the patients are described in Table 1. Seventy-three patients had a positive result for at least 1 paraneoplastic antibody, and 31 had a clinical diagnosis without a detectable paraneoplastic antibody. PET results were positive in 24 patients, 17 of whom had a positive result for at least 1 paraneoplastic antibody. Concomitant CT data were available for 71 patients. Results from CT were positive in 26 patients. A total of 31 patients underwent biopsy or surgery, and the rest were monitored (Figure 1). Ten patients had histologic or surgical proof of malignancy. Characteristics of all 10 patients are listed in Table 2. PET was able to detect 8 of 10 malignancies diagnosed (Figure 2). Computed tomography, by comparison, detected 3 of 10 malignancies (Figure 2). The 2 false-negative PET results were a 5-mm adenocarcinoma of the fallopian tube (Figure 3) and metastatic retroperitoneal lymph nodes
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
OCCULT MALIGNANCY IN SUSPECTED PNS
107 Patients underwent PET for suspected PNS
3 Did not meet inclusion criteria
104 Met inclusion criteria
73 PNS antibody positive
31 PNS antibody negative
54 PET negative
17 PET positive
7 PET positive
22 PET negative
1 Biopsy or surgery
0 Biopsy or surgery
2 Refused biopsy or surgery 15 Biopsy or surgery
15 Biopsy or surgery
FIGURE 1. Schematic chart of patient distribution. PET = positron emission tomography; PNS = paraneoplastic neurologic syndrome.
TABLE 2. Ten Patients With PNS Diagnosed With Malignancya Age (y)/ sex 36/M
Morvan chorea
83/F
PNS antibody
PET result
CT result
Final diagnosis
Follow-up (d)
Outcome
Positive
Positive
Thymoma
1227
Alive
Chorea
GAD-65 and potassium-channel antibody CRMP-5
Positive
Positive
1299
Dead
68/F
Chorea
ANNA-1
Positive
Negative
2069
Alive
74/F
Cerebellar degeneration Ganglioneuropathy
PCA-1
Negative
Negative
1037
Dead
ANNA-1 and N-type calcium channel ANNA-1
Positive
Negative
Adenocarcinoma of right lower lobe; small cell carcinoma of mediastinal nodes Mixed small and large cell carcinoma Serous carcinoma of right fallopian tube Small cell carcinoma
22
Dead
Positive
Positive
Small cell carcinoma
235
Positive
Negative
Small cell carcinoma
56
Positive Positive
Negative Negative
Negative
Negative
Small cell carcinoma Adenocarcinoma of unknown origin Metastatic spindle cell carcinoma diagnosed at autopsy
64/M 46/F 66/F 64/F 51/F 45/F
a
PNS
Peripheral polyradiculopathy Large-fiber sensory neuropathy Sensory myelopathy Multiple cranial nerve deficit Progressive multifocal neurological deficit
CRMP-5, PCA-2, and antiamphiphysin CRMP-5 ANNA-1 ANNA-1 and potassium-channel antibody
Lost to follow-up Dead
790 751
Alive Alive
50
Dead
ANNA = antineuronal antibody; CRMP = collapsin response-mediator protein; CT = computed tomography; GAD = glutamic acid decarboxylase; PCA = Purkinje-cell cytoplasmic autoantibody; PET = positron emission tomography; PNS = paraneoplastic neurologic syndrome. Mayo Clin Proc.
•
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
919
OCCULT MALIGNANCY IN SUSPECTED PNS
12
Malignancies diagnosed
10 8
6
4 2
0
Total malignancies
PET results positive
CT results positive
FIGURE 2. Distribution of malignancies and positive results from positron emission tomography (PET) and computed tomography (CT).
FIGURE 3. A 74-year-old woman with paraneoplastic neurologic syndrome. Computed tomography and fludeoxyglucose F 18 used with positron emission tomography demonstrated no hypermetabolic focus indicating malignancy. Patient had serous carcinoma of the right fallopian tube.
in a patient with previously resected spindle cell carcinoma of the uterus. Figure 4 represents positive PET results in a patient with an initial negative CT scan. Two patients with
positive results from PET deferred biopsy for unknown reasons and were lost to follow-up. Patients were followed up in a neurology clinic for assessment and treatment of
FIGURE 4. A 64-year-old woman with paraneoplastic neurologic syndrome. Computed tomography and fludeoxyglucose F 18 used with positron emission tomography demonstrated hypermetabolic mediastinal lymphadenopathy (arrows), which was proven to be small cell carcinoma via biopsy.
920
Mayo Clin Proc.
•
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
OCCULT MALIGNANCY IN SUSPECTED PNS
TABLE 3. Estimates of Sensitivity, Specificity, PPV, and NPV of PET and CTa
Sensitivity Specificity PPV NPV a
PET % (95% CI)
CT % (95% CI)
80 (44-98) 67 (43-85) 53 (27-79) 88 (62-99)
30 (7-65) 71 (42-89) 33 (8-70) 68 (45-86)
CI = confidence interval; CT = computed tomography; NPV = negative predictive value; PET = positron emission tomography; PPV = positive predictive value.
their neurologic disease. The median length of follow-up was 337 days (range, 22-2519 days). During the follow-up period, one patient with initial negative results from PET and an antibody result positive for PNS was diagnosed with small cell carcinoma of the lung on a follow-up PET study after 27 months. There were 16 false-positive results from PET. The specificity of PET was 67%, and the PPV was 53%. The sensitivity of PET for malignancy was 80%, and the NPV was 88% (Table 3). The sensitivity, specificity, PPV, and NPV of CT were 30%, 71%, 33%, and 68%, respectively. PET was significantly better than CT for detecting occult malignancy in our patient population (P=.02). DISCUSSION Patients with suspected PNS present a diagnostic dilemma because of the difficulty in detecting a suspected occult malignancy with conventional testing. Although PET has been recommended and used to detect occult malignancies in this patient population, its diagnostic utility has not been well defined.12 We have scrutinized the clinical utility of PET in patients with suspected PNS. Data suggest that PET has a better NPV (88%) and therefore is a better diagnostic tool to rule out occult malignancy, especially of the lung and mediastinum (PET did not miss any of these malignancies). All patients with cancer had a positive test for PNS antibody, and no patients with antibody-negative results were found to have malignancy. If we consider only antibody-positive patients, the diagnostic utility of PET scan is further enhanced. Data also revealed a high false-positive rate of PET, necessitating confirmation with a tissue diagnosis. The clinical dilemma remains for patients with positive antibody results and no cancer detected with PET.17,18 In our study, 62 of 73 antibody-positive PNS patients had no malignancy detected after extensive work-up and follow-up. There is a remote possibility that these patients might have early malignancy, and development of autoantibodies might have resulted in resolution of the tumor Mayo Clin Proc.
•
or rendered it indictable because of its size.13 The Paraneoplastic Neurological Syndrome Euronetwork recommends a follow-up period of 5 years to declare that a patient does not have any occult malignancies.14 We observed only one malignancy in these patients during follow-up. Thus, it can be inferred that not all patients with a positive test for paraneoplastic antibody and PNS will have occult malignancies. Our study is limited by its retrospective nature, the small patient population, and the rarity of this condition. Nevertheless, it remains one of the largest evaluations of the utility of PET for detecting occult malignancy. The high rate of false-positive results from PET and the inability of PET to detect smaller (8-10 mm) tumors limits the test’s value. Some patients in our study had limited follow-up at our institution because longdistance travel and neurologic impairment made return visits difficult. CONCLUSION PET is substantially more sensitive than CT for detecting occult malignancy in patients with suspected PNS. Occult malignancy was detected only in patients with PNS who had a positive test for autoantibody. Patients with suspected PNS and no detectable malignancy should be followed up every 6 to 12 months for a period of at least 5 years as recommended by the European Federation of Neurological Societies. The utility of follow-up CT or PET in these patients could not be determined because of limited data.
REFERENCES 1. Darnell RB, Posner JB. Paraneoplastic syndromes involving the nervous system. N Engl J Med. 2003;349(16):1543-1554. 2. Darnell RB. Onconeural antigens and the paraneoplastic neurologic disorders: at the intersection of cancer, immunity, and the brain. Proc Natl Acad Sci U S A. 1996;93(10):4529-4536. 3. Pittock SJ, Kryzer TJ, Lennon VA. Paraneoplastic antibodies coexist and predict cancer, not neurological syndrome. Ann Neurol. 2004;56(5):715-719. 4. Bredholt G, Storstein A, Haugen M, et al. Detection of autoantibodies to the BTB-kelch protein KLHL7 in cancer sera. Scand J Immunol. 2006;64(3):325-335. 5. Anderson NE, Rosenblum MK, Graus F, Wiley RG, Posner JB. Autoantibodies in paraneoplastic syndromes associated with small-cell lung cancer. Neurology. 1988;38(9):1391-1398. 6. Moll JW, Antoine JC, Brashear HR, et al. Guidelines on the detection of paraneoplastic anti-neuronal-specific antibodies: report from the Workshop to the Fourth Meeting of the International Society of Neuro-Immunology on paraneoplastic neurological disease, held October 22-23, 1994, in Rotterdam, The Netherlands. Neurology. 1995;45(10):1937-1941. 7. Darnell RB. Paraneoplastic neurologic disorders: windows into neuronal function and tumor immunity. Arch Neurol. 2004;61(1):30-32. 8. Albert ML, Darnell RB. Paraneoplastic neurological degenerations: keys to tumour immunity. Nat Rev Cancer. 2004;4(1):36-44.
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
921
OCCULT MALIGNANCY IN SUSPECTED PNS
9. Nardelli E, Curro-Dossi B, Rizzuto N, Graus F. Autoimmune paraneoplastic cerebellar degeneration: immunohistological localization of antibody-binding sites. Clin Neurol Neurosurg. 1995;97(1):89-94. 10. Graus F, Lang B, Pozo-Rosich P, Saiz A, Casamitjana R, Vincent A. P/Q type calcium-channel antibodies in paraneoplastic cerebellar degeneration with lung cancer. Neurology. 2002;59(5):764-766. 11. Rojas-Marcos I, Rousseau A, Keime-Guibert F, et al. Spectrum of paraneoplastic neurologic disorders in women with breast and gynecologic cancer. Medicine (Baltimore). 2003;82(3):216-223. 12. Voltz R, Graus F. Diagnosis and treatment of paraneoplastic neurological disorders. Onkologie. 2004;27(3):253-258. 13. Graus F, Keime-Guibert F, Rene R, et al. Anti-Hu-associated paraneoplastic encephalomyelitis: analysis of 200 patients. Brain. 2001;124(pt 6): 1138-1148. 14. Vedeler CA, Antoine JC, Giometto B, et al, Paraneoplastic Neurological Syndrome Euronetwork. Management of paraneoplastic neurological syndromes: report of an EFNS Task Force. Eur J Neurol. 2006;13(7):682-690. 15. Graus F, Delattre JY, Antoine JC, et al. Recommended diagnostic criteria for paraneoplastic neurological syndromes. J Neurol Neurosurg Psychiatry. 2004;75(8):1135-1140. 16. Bataller L, Graus F, Saiz A, Vilchez JJ, Spanish Opsoclonus-Myoclonus Study Group. Clinical outcome in adult onset idiopathic or paraneoplastic opsoclonus-myoclonus. Brain. 2001;124(pt 2):437-443. 17. Graus F, Bonaventura I, Uchuya M, et al. Indolent anti-Hu-associated paraneoplastic sensory neuropathy. Neurology. 1994;44(12):2258-2261. 18. Darnell RB, DeAngelis LM. Regression of small-cell lung carcinoma in patients with paraneoplastic neuronal antibodies. Lancet. 1993;341(8836):2122.
922
Mayo Clin Proc.
•
APPENDIX. Paraneoplastic Antibodies and Their Normal Valuesa Normal values at Mayo Clinic laboratories
Paraneoplastic antibodies AChR Binding Ganglionic N ANNA type 1 (anti-Hu) 2 (anti-Ri) 3 Unclassified antineuronal
)0.02 nmol/L )0.02 nmol/L <1:240 titer <1:240 titer <1:240 titer Nonspecific IgG in titers higher than 1:240
PCA type 1(anti-Yo) 2 Tr Amphiphysin Collapsin response-mediator receptor-5 (anti-CV-2) Striated muscle N-type calcium channel P/Q-type calcium channel Potassium channel Glutamic acid decarboxylase 65 a
<1:240 titer <1:240 titer <1:240 titer <1:240 titer <1:240 titer <1:60 titer )0.02 nmol/L )0.02 nmol/L Absent <0.02 nmol/L
AChR = acetylcholine receptor; ANNA = antineuronal antibody; PCA = Purkinje-cell cytoplasmic autoantibody
August 2008;83(8):917-922
•
www.mayoclinicproceedings.com
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.