The incidence of patients with endometrial cancer diagnosed at the time of scraping/biopsy but no residual cancer found after hysterectomy has not been well studied in the past. The goal of this study was to investigate the “disappearance” of this cancer through long-term follow-up. All initial scalpel biopsies and subsequent hysterectomy specimens were re-read by a gynecologic pathologist and the diagnosis confirmed. Serial sections of all endometrium were made for histological examination. Clinical and pathological indicators including patient data, tumor histological type and grade, stage, biopsy modality, adjuvant chemotherapy, surgical modality, recurrence, and disease-specific survival rates were analyzed. A total of 23 patients with preoperative biopsy confirmed endometrial carcinoma and no residual cancer after hysterectomy were diagnosed, including 15 (65.2%) with endometrioid carcinoma, 6 (26%) with plasmacytoma, and 1 (4.3%) each with plasmacytoma and clear cell carcinoma. 17 patients underwent cervical dilatation and diagnostic scraping, and 6 had endometrial biopsy. The median follow-up time was 8.8 years (from 1.2 to 17 years). Only 2 patients with plasmacytoma underwent adjuvant chemotherapy, and no patient underwent radiotherapy. Only one patient died after 27 months with an initial diagnosis of FIGO grade II endometrioid carcinoma by dilated cervical scraping. The preoperative scraping/biopsy diagnosis of endometrial cancer and the failure to find residual cancer after hysterectomy cannot be attributed to technical pathology errors. The overall prognosis for patients whose cancer “disappears” is excellent, although there is no targeted “standard” treatment. In any case, the patient is recommended to be followed up closely.