Validation of heart failure prediction tool in cardio-oncology breast cancer population.
Bibliographic record
Abstract
e17693 Background: While advances in breast cancer treatment, including targeted therapies such as trastuzumab have improved patient outcomes, short and long-term cardio-toxicity is a recognized risk. A clinical risk score (CRS) based on clinical risk factors has been derived for breast cancer (BC) patients (Ezaz et al, 2014), but remains untested in a real world clinical population. The objectives of this study are to apply this CRS to a real world breast cancer (BC) population seen in a dedicated cardio-oncology referral clinic (CORC) and correlate predicted versus actual risk of congestive heart failure (CHF) and cardiomyopathy (CM). Methods: BC patients referred to the CORC between October 2008 and August 2014 were reviewed retrospectively. Data was collected on patient demographics, cardiac risk factors, cardiac testing and outcomes. A CRS was calculated for each patient. Sensitivity, specificity, positive and negative predictive values of this risk score were evaluated using CHF/CM as the end-point. Results: 337 BC patients were reviewed; 14 were excluded because of missing data. Median age was 56 years old (range 25 - 87, SD 12); 217 (66%) had early stage (I-II) disease; 214 (66%) were ER positive; 181 (56%) were PR positive; and 199 (62%) were Her2 positive. 93% (n = 301) received adjuvant chemotherapy and 63% (n = 203) received targeted agents. Applying the CRS found 194 (60%) would be considered low risk to develop CHF/CM; 78 (24%) moderate risk; and 51 (16%) high risk. When applied to this population, the high-risk score cut-off had a 30.2% sensitivity (CI 18.7 – 44.5%) and 87.0% specificity (CI 82.2 % - 90.6%). Positive predictive value is 0.314 (CI 0.195 – 0.460) and negative predictive value is 0.86 (CI 0.816 – 0.901). Conclusions: This CRS has modest positive predictive value and good negative predictive value for CHF and CM in a contemporary sample of breast cancer patients. Patients with a high CRS may benefit from more intense cardiac evaluation in a clinic such as our CORC, since almost half of these patients develop CHF or CM. Most patients have a low or moderate CRS, and a lower risk of CHF/CM. Future work combining clinical risk scores with cardiac imaging and biomarkers may better predict cardiac risk during breast cancer therapy.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.013 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".