Scientific Papers

Anti-Tr/DNER antibody paraneoplastic cerebellar degeneration preceding a very late relapse of Hodgkin Lymphoma after 12 years | Cerebellum & Ataxias

Our case is atypical in several ways. Normally, the PCD precedes the diagnosis of HL in 80 % of the patients [1, 2]. In our patient the recurrence occurred 12 years after a complete remission. Second, HL has a high cure rate and a late relapse is very rare [4]. If it occurs it usually present as symptomatic lymphadenopathy and/or B-symptoms (fever, night sweats, or unintended weight loss) [4]. Presentation of a very late relapse of HL as a PCD has to our knowledge not previously been reported. Third, the neurological outcome in our elderly patient was better than expected. Combining data from the 2 largest series of PCD in HL only 14 % had a full or a partial neurological recovery and patients younger than 40 years were more likely to improve [1, 2]. Fourth, the patient was admitted to your stroke unit where the correct diagnosis initially was not made, although PCD in rare cases can mimic a stroke within the posterior cerebrovascular territory [5].

Opposite, to the well-documented value of early identification of a PNS in patients with unknown malignancy, the importance of these syndromes in the detection of a cancer relapse is unknow and there are limited data on the frequency of PNS in patients with reoccurrence of their malignancy. In a case series of 28 patients with PCD in HL 5 patients had onset of PCD after their lymphoma [2], in 3 of these within the first year. One patient had a mild gait ataxia 10 years after the cure of HL, but no tumor was found after 4 years observation. Delayed onset of a second PNS has been described in 8 patients identified among 979 cases in the PNS Euronetwork database [6]. In 5 of these the second PNS episodes antedated a cancer relapse or a new cancer. In all patients, the second PNS was clinical different from the first, although the antibody was the same, most commonly Anti-Hu. The reason why our patient did not develop PCD, at the time of the initial diagnosis of HL remain unclear. At that time the tumor burden was significantly higher (Ann Arbor stage III B) than at the relapse 12 years later (stage 1 A). This observation favors the idea that the immune response was more aggressive at the relapse but at the same time more effective in controlling the cancer [6] and is supported by the finding that in SCLC the presence of anti-Hu antibodies correlated with limited tumor stage, response to chemotherapy and longer survival [7].

The pathological hallmark of PCD is a widespread loss of Purkinje cells of the cerebellar cortex. The target antigen of the Tr-antibodies has been identified as DNER (Delta/notch-like epidermal growth factor-related receptor) [8, 9]. This protein is strongly expressed in the dendrites of Purkinje cells and is important for the function of these neurons. Unlike other onconeuronal antibodies in PCD (e.g., anti-Hu in SCLC, anti-Yo in ovarian or breast cancer), there is no evidence of immune activity to anti-Tr/DNER antibodies in malignant lymphoma tissue suggesting, that the pathological immune response is not triggered by tumor expression itself but likely from a more complex immune dysregulation caused by the tumor [8]. In addition to anti-Tr antibodies, mGluR1 and mGluR5 antibodies can be associated with HL. Both these antibodies were negative in our patient. mGluR1 antibodies has been described in two patients with PCD 2 and 9 years after HL [10]. In contrast to anti-Tr and mGluR1 antibodies, mGluR5 antibodies associated with HL causes a subacute syndrome with confusion and neuropsychiatric symptoms [11].

PCD related to the Tr/DNER antibodies does not respond as well to treatment as other disorders associated to antibodies against cell surface antigens does. Various treatment strategies (high-dose steroids, IVIG and plasmapheresis given a single therapy or in combination) have been reported. Our patient had benefit of serial IVIG treatments in combination with promptly cancer treatment.

In our patient there was a diagnostic delay of 3–4 weeks before a correct diagnosis was made and the case demonstrates three important diagnostic pitfalls: (1) Initially, the symptoms were wrongly interpreted as a stroke, (2) HL is normally considered as cured after 5 years without a relapse. The FDG-PET/CT finding could easily had been misdiagnosed as a Warthin tumor which is a benign condition so that further investigation not was initiated and the relapse had remained undiagnosed and (3) lack of attention of the oncological history at the first admission. This highlights that the presence of a paraneoplastic syndrome should be considered even in patients with very long cancer remission so that treatment can be initiated as early as possible. The increasing high cure rate and low relapse rate in many cancers does not exclude the possibility that a PNS can be the first and only sign of a relapse.

Source link