Malignant teratoma in Klippel–Feil syndrome: a case report and review of the literature

The association between posterior fossa dermoid tumor and Klippel–Feil syndrome is well known. However, the association of cervical fusion anomalies with malignant teratomatous lesions has rarely been reported.

Many hypotheses have been proposed to explain the embryological association of the two pathological conditions. One theory suggests that failure of the segmentation of the cervical sclerotomes could lead to the Klippel–Feil anomaly. In accordance with this, the absence of the cleavage of the ectoderm from the neuroectoderm would result in entrapment of the dermal elements within the closing neurotube [11]. Another theory suggests that during the formation of the cephalic and cervical brain flexures, a shortening of the cervical spine related to the fusion of the somites, may result in altered tissue tension. This process would lead to the entrapment of immature dermal elements [16]. Following the neural tube overdistension and somites distortion, the expression of Hox and Pax genes, the DNA sequences that control the development of intervertebral disk may be reduced producing the vertebral segmentation. It has also been speculated that this occurrence, during the 28th to 35th days of life, may cause an altered tissue tension at the cervicomedullary junction resulting in the entrapment of dermal elements. This process may be responsible for the formation of posterior fossa dermoid tumors [10, 16].

There are different mechanisms underlying the migration of germ cells including pseudopodial structures, chemotactic factors, basement membrane or extracellular molecules [16].

Another important structure involved in posterior fossa embryological development is the isthmus; the isthmus develops at the junction of the mesencephalon and metencephalon, thus serving as an organizing center for both midbrain and hindbrain differentiation. The homeobox genes play a crucial role in segmentation and subsequent patterning. Among them, the Sonic hedgehog regulates the midbrain–hindbrain morphogenesis through positive regulation of the Gli activators (GLI1) and inhibition of the Gli repressors (GLI3) and controls the overall growth of this region. Sonic hedgehog also restricts FGF8 expression to the isthmus, which is essential for the differentiation of the tecto-isthmo-cerebellar region [17]. The encoded protein SNF5 has an oncosuppressor role by cooperating with p53 in the inhibition of GLI1; the loss of SNF5 causes the aberrant activation of the Sonic hedgehog pathway and drives teratoid tumorigenesis through the expression of GLI1, as documented by the growth of SNF5-deficient malignant rhabdoid cells in vivo and in vitro [18]. Sonic hedgehog and GLI1 are also crucially involved in the morphogenesis of the tecto-cerebellar midline structures, as previously mentioned. The correlation between embryogenesis and tumorigenesis is well characterized by the involvement of Sonic hedgehog signaling in medulloblastoma [18].

In our case we report an early tumor regrowth, after a subtotal removal, due to tight adherences of the tumor to the floor of the fourth ventricle. The extremely aggressive features of this kind of lesion have to be related to the nature of teratomatous tumors. The so-called “growing teratoma syndrome” is a rare complication which follows the resection of a recurrent intracranial nongerminomatous germ cell tumor in adults [19]. It is observed after partial response to multimodality therapy and despite a decrease in tumor serum markers. In these cases, the enlarging tumors consist of elements of mature teratoma that presumably are refractory to chemotherapy or radiation, and are selected by these same modalities of treatment. In our case, the tumor regrowth was caused by the surgical treatment, due to the removal of the less aggressive intraventricular tumor leaving the most aggressive portions of the tumor, which proliferate very quickly.