Gestational trophoblastic diseases: 2. Hyperglycosylated hCG as a reliable marker of active neoplasia.

Objectives: To determine whether circulating hyperglycosylated human chorionic gonadotropin (hCG-H), a promoter of choriocarcinoma growth and tumorigenesis, is a reliable marker of active gestational trophoblastic neoplasia (GTN) or choriocarcinoma, and whether hCG-H can consistently discriminate quiescent gestational trophoblastic disease (GTD) from neoplasia.
Methods: Patients were those referred to the USA hCG Reference Service for consultation. These included a total of 82 women with GTN, including 30 with histologic choriocarcinoma. They were compared with 26 patients with resolving hydatidiform mole and 69 with quiescent GTD (persistent positive low value of real hCG but no clinical evidence of disease). All were tested for total hCG and hCG-H. hCG-H was calculated as the percentage of total hCG (hCG-H(%)).
Results: We compared the utility of total hCG and hCG-H(%) in detecting active GTN and quiescent GTD. There was no significant difference when measuring total hCG (includes regular and hyperglycosylated hCG), between women with quiescent GTD and self-resolving hydatidiform mole compared to choriocarcinoma/GTN cases (P > 0.05 and P > 0.05). In contrast, hCG-H(%) was significantly higher in choriocarcinoma/GTN cases (P < 0.000001, and P < 0.000001). The usefulness of hCG and hCG-H(%) testing was assessed for discriminating between the 69 quiescent GTD cases, which required no therapy, and choriocarcinoma/GTN which need treatment. While hCG would detect 62% and 24% of malignancies at a 5% false positive rate, hCG-H(%) would detect 100% and 84% of malignancies at this same false positive rate. Follow-up data were received and repeat consultations were performed in 23 cases in which active disease was subsequently demonstrated. In 12 of 23 cases, hCG-H(%) results were able to first identify active disease 0.5 to 11 months prior to rapidly rising hCG or detection of clinically active neoplasia. In the remaining 11 cases, hCG-H(%) active disease appeared at the same time as rising hCG or demonstrable clinical tumor.
Discussion and Conclusion: hCG-H(%) appears to reliably identify active trophoblastic malignancy. It is a 100% sensitive marker for discriminating quiescent GTD from active GTN/choriocarcinoma. It is also a marker for the early detection of new or recurrent GTN/choriocarcinoma. The data presented appear sufficient to encourage the adoption of hCG-H as a tumor marker in trophoblastic disease. Further studies are now urgently required to confirm and extend our findings.