Virtual Library
Start Your Search
Yuxuan Pan
Author of
-
+
P2.12 - Small Cell Lung Cancer/NET (ID 180)
- Event: WCLC 2019
- Type: Poster Viewing in the Exhibit Hall
- Track: Small Cell Lung Cancer/NET
- Presentations: 1
- Now Available
- Moderators:
- Coordinates: 9/09/2019, 10:15 - 18:15, Exhibit Hall
-
+
P2.12-22 - Risk Factors for BM Incidence in SCLC: A Predictive Model for SCLC Patients on Brain Metastasis (Now Available) (ID 1299)
10:15 - 18:15 | Author(s): Yuxuan Pan
- Abstract
Background
As one of the serious complications in patients with lung cancer, brain metastasis (BM) have been demonstrated more common when patients were diagnosing with small-cell-lung cancer (SCLC). After developing a BM, patients’ median overall survival will be obviously shorten. Prophylactic Cranial Irradiation (PCI) used to be an optimal therapy to prevent the occurrence of BM in SCLC patients, while recently its status was shaken. This retrospective study aimed to find out the potential risk factor relative to the incidence of BM and construct a predictive mode for SCLC, also to see the efficiency of PCI on SCLC patients’ ICPFS.
Method
Patients pathologically diagnosed with SCLC were consecutively enrolled from May 1st 2006 to February 15th 2019 and reviewed. The intracranial progression-free survival (ICPFS) were calculated by a Kaplan-Meier approach. And the candidate prognosticators of the ICPFS were checked by COX regression analysis. Using these risk factor and its coefficient obtained in the multivariate analysis , a model were constructed to anticipate the incidence of BM. And the receiver-operating characteristic (ROC) curve was constructed to access the prediction ability.
Result
Totally 261 SCLC patients were eligible for analysis, including 78 cases had developed a BM (29.9%) after the first diagnosis. We used Cox proportional hazards to analyze and adjust the confounding factor to the incidence of BM, and got the result, that there were five factors showed the significance both in Uni-(P<0.15) and multivariate(P<0.05) analysis, including smoking index <400(P=0.017); vascular invasion (P=0.012); number of primary lesion (P=0.042); staging (P=0.018); and hyponatremia (P=0.05). And a predicative model was constructed by using these factors (except the hyponatremia), with an AUC value of 0.743. And we used this model to distinguish low and high risk group. Two groups’ ICPFS showed a remarkable significance (P<0.001). We also simplified it to make it more convenient in clinical work, which also showed an outstanding significance (P<0.001). However, the PCI in our study was not reach the significance, even though it prolonged the ICPFS (P=0.211).
Conclusion
Factors included smoking index >=400, vascular invasion, number of primary lesion, staging, and hyponatremia, increasing the risk of BM occurrence. In the era of individualized treatment, a predictive model based on smoking status and image science can give some suggestions for clinical work by anticipating the highly risky incidence of brain metastasis on SCLC patients.