[Classification of pancreatic neuroendocrine tumours: Changes made in the 2017 WHO classification of tumours of endocrine organs and perspectives for the future]

The WHO classification of the tumors of endocrine organs, published in July 2017, has introduced significant changes in the classification of pancreatic neuroendocrine tumors, the previous version of which has appeared in 2010, within the WHO classification of the tumors of the digestive system. The main change is the introduction of a new category of well-differentiated neoplasms, neuroendocrine tumors G3, in addition to the previous categories of neuroendocrine tumors G1 and G2. The differential diagnosis between neuroendocrine tumors G3 (well-differentiated) and neuroendocrine carcinomas (poorly-differentiated) might be difficult; the authors of the WHO classification therefore suggest the use of a number of immunohistochemical markers to facilitate the distinction between the two entities. The other changes are: (a) the modification of the threshold between neuroendocrine tumors G1 and G2, now set at 3%; (b) the terminology used for mixed tumors: the previous term mixed adeno-neuroendocrine carcinoma (MANEC) is substituted by the term mixed neuroendocrine-non neuroendocrine neoplasm (MiNEN). Finally, the recommendations for Ki-67 index evaluation are actualized. Even if these changes only concern, stricto sensu, the neuroendocrine tumors of pancreatic location, they will probably be applied, de facto, for all digestive neuroendocrine tumors. The revision of the histological classification of pancreatic neuroendocrine tumors coincides with the revision of their UICC TNM staging; significant changes have been made in the criteria for T3 and T4 stages. Our professional practices have to take into account all these modifications.

[Intramedullary ependymomas: A French retrospective multicenter study of 221 cases]

AIM

Intramedullary ependymomas (IE) are the most frequent intramedullary tumors in the adult population. The gold standard treatment is to obtain gross total removal without any new postoperative neurological deficit. The authors report the results of a multicenter series with a long-term follow-up and the analysis of clinical, radiological, surgical data as well as the complementary treatments.

MATERIALS AND METHODS

A retrospective analysis of adult patients with IE operated on between January 1984 and December 2011 at 7 French centers (Bordeaux, Kremlin-Bicêtre, Lille, Lyon, Marseille, Montpellier, Nice, and Nîmes) was performed. The minimal follow-up was 12 months. The clinical evaluation was based on the McCormick classification in the pre and postoperative period at 3 months, 1 and 5 years.

RESULTS

Data of 221 adult patients with a pathologically confirmed diagnosis of IE were considered: 134 patients were treated at the Neurosurgical Department of Kremlin-Bicêtre Hospital, 26 were treated at Lille and 61 were treated in the southern region of France (Marseille, Bordeaux, Montpellier, and Lyon). The epidemiological analysis was performed on the entire cohort of patients, while follow-up considerations were made solely on the 134 patients managed at Kremlin-Bicêtre Hospital to obtain homogeneous data. A slight male prevalence was observed (59 % of cases), with an average age of 41.8 years at diagnosis. The mean age at first clinical manifestations was 39.6 years, thus the average duration of symptoms before the diagnosis was 29 months. Neuropathic pain and neurological deficit were the most revealing symptoms in 64 % and 32 % of cases respectively. The localization of the IE was basically cervical in 35.7 %, primarily thoracic in 25.8 % and cervico-thoracic in 22.2 %. The mean tumor length in the sagittal plan was 20.4mm (range 1 to 99mm). A cystic cavity was present in 76.5 % of cases while an intratumoral hemorrhage was detected in 30.3 % of cases. The rate of complete removal was performed in 79.2 % of cases when considering the whole cohort and in 91 % of cases treated at Bicêtre Hospital. In 95 % of cases a WHO grade I or II ependymoma was isolated and in 5 % of cases a WHO grade III. The McCormick scale (MCs) (Lou et al., 2012) [1] was used to rate the degree of preoperative functional impairment in 4 grades. In the immediate postoperative period a worsening of functional capacity was observed. Only 28.8 % of patients had MCs 1 in the immediate postoperative period; 34.8 % had MCs 2; 20.4 % had MCs 3 and 16 % had MCs 4. At 5 years of follow-up (101 patients) the frequencies of grades 1 and 2 were increased: 59 % of patients had MCs 1, 20.8 % had MCs 2, 10,9 % had MCs 3 and 9.3 % had MCs 4. The extension of the lesion on the sagittal plan calculated on the preoperative MRI, was the only predictive factor associated with the immediate postoperative outcome and the short-term follow-up (P=0.04), whereas the preoperative neurological status is the only predictive factor for long-term follow-up (P=0.005).

CONCLUSION

Gross total removal remains the mainstay treatment for IE. Early surgery is indicated if the patient is symptomatic or the tumor increases in size. A postoperative regular follow-up is mandatory for at least 10 years due to the risk of recurrence. If a growing residue is detected, a second intervention is recommended without any adjuvant treatment if a WHO grade I lesion is confirmed by the pathological analysis. Complementary treatment should be reserved for high-grade ependymomas or in case of unresectable and progressive residue.

[Pertinence of the preoperative exploration in the evaluation of the risk of lymph node metastasis in endometrial cancer]

PURPOSE

To assess the relevance of MRI, endometrial biopsy and curettage in the diagnosis of endometrial cancer at high risk of lymph node metastasis.

PATIENTS AND METHODS

A retrospective study on continuous series of patients treated for endometrial cancer limited to the uterus between 2004 and 2008, results of preoperative evaluation of tumor stage using MRI, histological type and grade by endometrial curettage and biopsies were compared to final histological examination.

RESULTS

One hundred and sixty-nine patients were included in the study. Ninety (53.3%) had MRI, 112 (66.2%) curettage and 61 (36.6%) endometrial biopsy using Pipelle de Cornier. Sensibility (SN), specificity (SP), positive (PPV) and negative predictive values (NPV) of MRI, in the diagnosis of endometrial cancer at high risk of lymph nodes metastases were of 65.6%, 87.2%, 77.7%, 79.2%. For EB and curettage SN, SP, PPV and NPV were of 42.9%, 96.9%, 85%, 79.5%; 80.6%, 98.3%, 96.2% and 90.6% respectively. 37.8% of cancers diagnosed to be at low risk of lymph node metastasis were at high risk in definitive histologic examination.

DISCUSSION AND CONCLUSION

Preoperative evaluation by MRI, endometrial curettage and biopsy has good diagnostic value in the identification of endometrial cancer susceptible to benefit from lymphadenectomy. Underestimation, however, is encountered in approximately one third of cases.