Page 74 - Read Online

Page 70 Shi et al. Neuroimmunol Neuroinflammation 2020;7:68-72 I http://dx.doi.org/10.20517/2347-8659.2019.16

A B

C D

Figure 1. Chest CT scans and pathological result. A: only a right lower lobe inflammatory focus can be found. The chest CT scan was
performed at the same time the eight cancer-driven mutations were detected (May 2017); B: the enlarged lymph nodes under the aortic
arch can be seen, and a shadow of soft tissue around the aortic arch of the left lower lobe probably reflects a malignant lesion (January
2018); C, D: morphological results reflect the possibility of small cell lung cancer. Scale bar: 400 µm (C); and 200 µm (D)

DISCUSSION
Given that PCD is associated with uncommon neuroendocrine tumors, such as SCLC, breast cancer, and
prostate cancer, the possibility of malignancy in patients with a subacute cerebellar syndrome should be
considered and excluded by clinicians. Several antineuronal antibodies are related to PCD, and more than
90% of patients harboring those antibodies are confirmed to have a tumor, mostly SCLC . In patients with
[10]
positive anti-CV2/CRMP5 antibodies, cerebellar ataxia, uveitis, and peripheral neuropathy are the most
commonly occurring symptoms . CRMP5, a protein belonging to a family of developmentally regulated
[6]
neural proteins, expresses in regions of the central nervous system undergoing postnatal plasticity or in
[10]
the peripheral nervous system mediating Schwann cell differentiation and axon repair . In both anti-
Hu and anti-CV2/CRMP5 antibodies positive patients, SCLC was the most commonly found cancer, while
[11]
malignant thymoma was found in patients only positive for anti-CV2 antibodies . However, considering
[6]
that the detection of those antibodies has very low sensitivity and specificity , clinicians should be more
vigilant, especially in cases with initial negative clinical examining results but with high suspicion of
malignancy.

Together with epigenetic and transcriptomic alterations, cancer is primarily driven by the accumulation
of somatic mutations in the genome over one’s lifetime. In recent years, the identification of somatic
mutations in cancer genomes has been revolutionized by high-throughput DNA sequencing and all types
of somatic mutations can be revealed by whole-genome sequencing. In light of these advances, precision
medicine and precision oncology have become possible, and treatments tailored based on an individual’s
mutational profile could be made.

In this case, no abnormalities were found via imaging or CSF examination until morphological and
immunohistochemical results supported the diagnosis of SCLC, seven months after eight somatic mutations
were detected. To be specific, we collected 278 somatic cancer-driven mutations from a cancer database
and used target region sequencing to find candidate mutations on this occasion. Mutation frequency,
variation prediction, and other analytic methods were utilized to analyze the data, and eight high-
frequency gene mutations were identified in our patient: TSC2, DNMT1, CIC, FGF6, NSD1, TSHR, CRLF2,