When and how is it appropriate to search for an underlying tumor?

In seropositive autoimmune encephalitides, the probability and most common localization of an underlying tumor can be predicted regardless of the detected neuronal antibody (Table 1). Of course, risk factors (e.g. smoking history), age and gender should also be taken into account in these considerations. Tumor diagnostics should be graded (Table 2).

Tabelle 1: Association between cancer and the most common antineuronal antibodies. SCLC = Small cell lung cancer, LEMS = Lambert-Eaton myasthenic syndrome, DNER = Delta/notch-like epidermal growth factor-related receptor, *in some patients Ma antibodies coexist, meaning that brainstem syndromes and non-testicular tumors will predominate.


Most common cancers

Antibodies against intracellular antigens (onconeural antigens). Cancer-related >95%

Anti-Hu (ANNA-1)

Lung cancer (85%), mostly SCLC, neuroblastoma, Merkel cell carcinoma, other tumors with neuroendocrine differentiation

Anti-Yo (PCA-1)

Ovarian cancer, breast cancer


SCLC, thymoma


Testicular tumor

Anti-Ri (ANNA-2)

Breast cancer, ovarian cancer, SCLC


Breast cancer, SCLC



Anti-SOX-1 (AGNA)


Anti-Tr (PCA-Tr), DNER

Hodgkin's, non-Hodgkin's lymphoma


SCLC (idiopathic forms known, cancer-related <90%)

Antibodies against synaptic antigens (neuronal surface antigens). Variable cancer relationship

Anti-NMDA receptor

Age and gender dependent. Women aged between 12-45 years in Germany
Ovarian teratomas (25%)

Anti-AMPA receptor

SCLC, thymoma, breast cancer (50%)

Anti-GABA(B) receptor

SCLC (50%)


Thymoma (<5%)


Depending on the syndrome: Limbic encephalitis various tumors <5%; Morvan syndrome (40%)


Hodgkin’s lymphoma (approx. 50%)


B cell neoplasms (<10%)

Anti-GABA(A) receptor

Thymoma (30%)


Not known


No known cancer relationship

Modified after the Practice Guideline on “Paraneoplastic Syndromes” issued by the German Neurological Society (DGN)


Table 2: Graded tumor diagnostics guided by the presumed localization for select tumors. Sensitivity in parentheses, if known. EB-US = Endobronchial ultrasound

Cancer Diagnostics
Primary Secondary Tertiary
Lung cancer Chest CT (80-85%),
Chest MRI
FDG-PET or FDG-PET/CT Bronchoscopy/EB-US and, if indicated, fine-needle aspiration
Mediastinoscopy, if indicated
Thymoma Chest CT (75-90%),
Chest MRI
Breast cancer Mammography (80%),
Breast MRI  
Ovarian cancer Transvag. ultrasound (69-90%) + CA-125 Pelvic/abdominal CT FDG-PET or
Ovarian teratoma Transvag. ultrasound (69-90%) MRI (93-98%) Chest CT (extrapelvic teratomas)
Testicular tumors Ultrasound (72%) + β-HCG, AFP Pelvic/abdominal CT (76%),
abdominal MRI
If indicated FDG-PET or FDG-PET/CT (malignant teratomas)
Lymphoma Chest CT/abdominal ultrasound FDG-PET or FDG-PET/CT  
Skin cancer (Merkel cell carcinoma) Dermatological examination
Biopsy, if applicable

Modified after the Practice Guideline on “Paraneoplastic Syndromes” issued by the German Neurological Society (DGN)



Find a center near you here.

Latest news

Aktuelle Impfempfehlung (SARS-CoV-2) des GENERATE Netzwerkes vom 07.05.2021

Upcoming events

Currently there are no events.


Brändle SM, Cerina M, Weber S et al. PNAS March 2, 2021 118 (9) e1916337118.

Dieses Paper wurde vom Research4Rare Verbund zum Paper of the Month gewählt und erscheint im April dieses Jahres im BMBF Newsletter!

Kreye J, Momsen Reincke S, Kornau HC et al.
Cell. 2020 Nov 12;183(4):1058-1069.e19.

Pitsch J, van Loo KMJ, Gallus M et al.
Ann Neurol. 2021; 00:1-20.

Hansen N, Schwing K, Önder D, et al. Epilepsy Behav. 2020;102:106682.

Bracher A, Alcalá C, Ferrer J, et al. Ann Clin Transl Neurol. 2020;7(2):239‐244.

Bien CG, Bien CI, Dogan Onugoren M, et al. [published online ahead of print, 2020 Apr 3] [published correction appears in J Neurol. 2020 May 12;:]. J Neurol. 2020;10.1007/s00415-020-09814-3.