Virtual Library
Start Your Search
Jeffrey Crawford
Author of
-
+
P2.01 - Advanced NSCLC (Not CME Accredited Session) (ID 950)
- Event: WCLC 2018
- Type: Poster Viewing in the Exhibit Hall
- Track:
- Presentations: 2
- Moderators:
- Coordinates: 9/25/2018, 16:45 - 18:00, Exhibit Hall
-
+
P2.01-35 - Predicting Risk of Chemotherapy-Induced Severe Neutropenia in Patients with Advanced Lung Cancer (ID 13500)
16:45 - 18:00 | Author(s): Jeffrey Crawford
- Abstract
Background
Neutropenia is associated with the risk of life-threatening infections, chemotherapy dose reductions and delays that may compromise treatment outcomes. The goal of this study was to develop simple prediction model for severe neutropenia in lung cancer.
a9ded1e5ce5d75814730bb4caaf49419 Method
A lung cancer dataset was assembled using data from existing national cooperative group phase II/III trials conducted between 1991-2010. Chemotherapy trials in patients with stages III and IV non-small cell lung cancer (NSCLC) or extensive small-cell lung cancer (SCLC) were included. We randomly selected 2/3 patients to derive the model, and the remaining were used for validation. Models were built with stepwise logistic regression and lasso regression on imputed data sets. We fitted the model on the imputed training data sets individually to get 10 models with 10 sets of selected predictors. Next we picked the union set and the intersection set of predictors from the models. The variables in the final model were selected by lasso regression, and then fitted into a logistic model. The performance of the model was evaluated by receiver operating characteristic (ROC) curve and the area under the ROC curve (AUC).
4c3880bb027f159e801041b1021e88e8 Result
The dataset was randomly separated into training [n=7606 (67%)] and testing sets [n=3746 (33%)]. The final predictive model included: Age (>65 years), gender (male), weight (kg), BMI, insurance status (yes/unknown), stage (IIIB/IV/ESSCLC), number of metastatic sites (1, 2 or ≥3), individual chemotherapy agents (gemcitabine, taxanes), number of chemotherapy agents (2 or ≥3), planned use of growth factors, associated radiation therapy, previous therapy (chemotherapy, radiation, surgery), duration of planned treatment, pleural effusion (yes/unknown), performance status (1, ≥2) and presence of symptoms (yes/unknown).
Figure: ROC Curve for Final model AUC=0.8306
8eea62084ca7e541d918e823422bd82e Conclusion
We have developed a relatively simple model with variables that are routinely available prior to treatment, to predict for neutropenia. This model should be validated prospectively.
6f8b794f3246b0c1e1780bb4d4d5dc53 -
+
P2.01-39 - Can Benefit or Futility in Treating Advanced Nsclc Be Determined Early Using the LCSS 3-Item Global Index (3-IGI) PRO? (ID 12299)
16:45 - 18:00 | Author(s): Jeffrey Crawford
- Abstract
Background
Background: Early assessment of the effect of treatment for advanced NSCLC can prevent unnecessary exposure to toxic and costly therapy while aiding in decision making to continue or change treatment. In a prior analysis in patients with mesothelioma (Symanowski JCO 2014), a 20% decline from baseline after 2 cycles of chemotherapy in the 3-Item Global Index of the LCSS identified patients unlikely to benefit. The 3-IGI (which evaluates: 1) global distress, 2) patient rated activities, and 3) quality of life, all in single VAS scales) takes less than 2 minutes to assess.
a9ded1e5ce5d75814730bb4caaf49419 Method
Methods: 164 patients with NSCLC receiving chemotherapy or checkpoint inhibitors were prospectively evaluated with the LCSS at baseline and every 3 weeks using electronic media. Patients were also randomized 1:1 so that their physicians knew the results of the LCSS immediately in half of the patients.
4c3880bb027f159e801041b1021e88e8 Result
Results: Patients: Stage IV 92%; first line 73%; female 43%; median PS 1; mean age 63. The LCSS was completed after 2 cycles of treatment and prior to planning for the next cycle (generally 6 weeks after baseline; representing 91% of the 148 patients living). Patients with a 20% decline in the 3-IGI compared with baseline had a median survival of 7.6 months, contrasted to 15.8 months for those without this degree of 3-IGI decline (p = 0.01); 1 year survivals = 26% versus 62%. Even with the marked PRO decline after 2 treatment cycles, patients in the 20% decline group received a median of 2.3 more cycles of the same chemotherapy (median cost = $10,712 per patient). In the 50% of patients for which their physicians knew the ongoing LCSS results, fewer chemotherapy and imaging studies were performed, but the differences were not significant (p = 0.8).
8eea62084ca7e541d918e823422bd82e Conclusion
Conclusions: Assessing change from baseline with the 3-IGI of the LCSS identifies after only 2 cycles of treatment those patients who have poor response and survival outcomes if continued on the same therapy. This PRO assessment is rapid, easy, and inexpensive. Physicians need to consider the impact of decline on decision options given that even when physicians were aware of the worsening PRO they often did not act on the findings. Patient responses to this validated PRO questionnaire provide valuable information that is not otherwise attainable. Responding to 3-IGI changes can result in better decisions concerning continuing or changing treatment, lessening toxicity, and savings in cost of unhelpful treatment.
Supported by: NIH/NCI R01 CA157409
6f8b794f3246b0c1e1780bb4d4d5dc53
