The state of the art management of germ cell tumors (GCT) in 2018 does not include novel agents targeting genomic alterations or exciting immunologic-based approaches but rather the avoidance of pitfalls in everyday practice. The relative rarity of GCT and high curability with correct management create the "perfect storm" for high-stakes errors to occur. This review focuses on several common pitfalls that should be avoided in staging and management of early-stage and advanced GCT in order to maximize patient outcomes. A particularly frequent misstep is to base treatment decisions on pre- rather than postorchiectomy tumor markers that, depending on marker directionality, can lead to either undertreatment with potentially inferior outcomes or overtreatment with excess toxicity. Another common mistake is the failure to consider the unique ability of GCT to differentiate and the distinct biology of teratoma (chemoresistance and lack of increased glucose uptake compared with normal tissue), which exerts a pervasive influence on nonseminoma management. This may lead to inappropriate use of PET scan to evaluate the postchemotherapy residual mass and, if negative, the conclusion that surgery is not needed whereas (FDG-negative) teratoma should be removed. It could also result in administration of additional unnecessary chemotherapy to patients with marker normalization but without robust radiographic response after 3 to 4 cycles of BEP. Finally, oncologists should strive to maintain standard chemotherapy doses, not substitute carboplatin for cisplatin, and refer to expert centers when expertise (e.g., RPLND) is not available locally in order to achieve optimal cure rates in advanced disease.
New polyp image classification technique using transfer learning of network-in-network structure in endoscopic images
