Multivariate Cox Regression Model Research Articles

Impact of human papillomavirus status on survival in patients with oral cancer.

To examine the association between the human papillomavirus (HPV) infection and overall survival rate in patients with oral cancer. This retrospective cohort study examined HPV status in 454 patients who were diagnosed with oral squamous cell carcinoma (OSCC) using the records of patients who underwent an initial treatment for OSCC between 2012 and 2021 at our institution as retrieved from the Cancer Registry database. The survival rates of the HPV-positive and HPV-negative groups were assessed and compared, and independent factors associated with survival were analyzed using multivariate Cox regression models. Of the 454 patients with OSCC included in this study, 73 were excluded for invalid HPV tests. Of the remaining patients, 39 and 342 patients were categorized into HPV-positive and HPV-negative groups, respectively. The prevalence of HPV-positive in the patients with OSCC was 10.2% (95% confidence interval 7.2%-13.2%). The 3-year overall survival rates were 56.2% and 53.9% in the HPV-positive and HPV-negative groups, respectively. The 3-year disease-specific survival rates in the HPV-positive and HPV-negative groups were 60.2% and 56.9%, respectively. The survival differences were not statistically significant. HPV-positive status was not a significant predictor of overall survival in the multivariable Cox regression analyses (p = 0.728). The prevalence of HPV-positivity among patients with OSCC in the study was 10.2%. No association was found between HPV-positive status and 3-year overall survival in patients with oral cancer. Level 3.

Retrospective analysis of factors impacting efficacy of patients with later stage breast cancer receiving neoadjuvant therapy.

e12624 Background: Neoadjuvant treatment is the current standard of care for patients with early breast cancer (EBC), especially for patients with later stage breast cancer (LBC). We conducted this retrospective study to summarize the correlation between pathologic complete response (pCR), disease-free survival (DFS), overall survival (OS) and clinical impacting factors of patients with later stage received neoadjuvant treatment. Methods: We retrospectively analyzed the clinical data of patients receiving neoadjuvant treatment at the Department of Breast Oncology, Peking University Cancer Hospital from Jan. 2013 to Feb. 2022. Statistical analyses were carried out using IBM SPSS Statistics. We calculated pCR rate, DFS and OS. Pearson’s χ2 or Fisher’s exact tests were used for comparison of categorical variables. DFS and OS were estimated using the Kaplan-Meier method and compared using the log-rank test. Cox regression was used to identify factors impacting on survival. All p values were two sided and p<0.05 was considered to denote statistical significance. Results: Of 198 patients, most of them were in stage II (115/198, 58.1%), followed by stage III (79/198, 39.9%) and stage I (4/198, 2.0%), The pCR rates for different molecular subtypes were 31.6% TNBC, 29.7% for HR-Her2+ BC, 26.1% for HR+Her2+ BC and 1.3% for HR+Her2- BC, respectively (p=0.000). The median DFS of patients with or without pCR was 59.63 months (95%CI: NA-NA) and 29.83 months (95%CI: 21.31-38.35), respectively (p=0.012). DFS of patients who were cT1 [59.63 months (95%CI: 37.95-81.31)] or cT2 [39.56 months (95%CI: 24.19-54.93)] was significantly longer than that of patients who were cT3-4 [25.76 months (95%CI: 14.11-37.41), p=0.041]. Miller Payne (M-P) grade and ypTNM stage were also significantly associated with DFS in univariate analysis and in the multivariate cox regression model after adjusting for molecular subtypes. The 5-year OS were 87.0% for patients who reached pCR and 52.0% for patients who didn’t (p=0.009). Conclusions: The pCR was significantly associated with molecular subtypes. HR negativity and overexpression of Her2 were significantly associated with higher pCR rate. Patients reached pCR had longer DFS and OS. DFS was significantly associated with cT stage (T1 vs T2 vs T 3-4), ypTNM stage and M-P grade.

Real world outcomes in patients with follicular lymphoma treated with RCHOP vs BR.

e19038 Background: In patients (pts) with (w) follicular lymphoma (FL), BR is associated w improved progression free survival (PFS) compared to RCHOP. In clinical practice, RCHOP is frequently utilized in pts w grade (G) 1-2 FL w high SUVs or Ki-67% proliferation index, and G 3A FL due to concern for transformation. Using real world data, we compared treatment trends and outcomes in pts w FL treated w first line (1L) BR versus (vs) RCHOP. Methods: This study used the nationwide Flatiron Health electronic health record-derived de-identified database. Pts diagnosed w FL G 1-2 or 3A and treated between (bt) 2011-2022 w 1L BR or RCHOP were included. The primary objective was to compare time to next treatment or death (TTNT) bt the two cohorts. Overall survival (OS), and transformation and POD24 rate were also analyzed. Univariable and multivariable Cox regression models were used to evaluate the impact of various clinical factors (sex, ethnicity, ECOG PS, treatment center type, FL G, FLIPI, rituximab maintenance [mtn]) on TTNT and OS. Results: The study included 2,089 pts; 1,475 pts received BR (BR n=1387, BO [obinutuzumab] n=88,), and 614 pts received RCHOP (RCHOP n=459, OCHOP n=8, RCEOP [etoposide] n=4, RCHP n=1, RCEP n=1, RCVP n=137, OCVP n=4). Median age at diagnosis was 64. Most patients were male (53%), non-Hispanic (90%), had advanced stage disease (82%), ECOG PS 0 (57%), G 1-2 FL (85%), intermediate FLIPI (62%), were treated at a community practice (83%), and did not receive rituximab mtn (54%). At median follow up of 90 mos, median TTNT bt BR and RCHOP was 96 vs 78 mos (HR 1.146, 95% CI 0.986-1.332, p = 0.076). In a multivariable model, there was no significant association bt RCHOP and TTNT. Male sex, ECOG PS ≥ 2, and high FLIPI were associated w worse TTNT, and Rituximab mtn w improved TTNT. Among G 3A pts (n=304), more pts received RCHOP (57% vs 43%), and median TTNT bt BR and RCHOP was 73 vs 78 mos (HR 0.787, 95% CI 0.542-1.142, p=0.207). Median OS bt BR vs RCHOP was 126 vs 138 months (HR 0.878, 95% CI 0.718-1.075, p=0.207). Male sex, ECOG PS ≥ 2, high FLIPI, and treatment in community setting were associated w worse OS. Rituximab mtn was associated w improved OS. 115 (5.5%) pts had transformation to aggressive lymphoma. Transformation rates bt BR vs RCHOP were 4.83% vs 6.50% (p=0.2) among all G 1-2 pts (n=1785), and 3.85% vs 9.2% (p=0.069) among all G 3A pts (n=304). 310 (14.8%) pts had POD 24, including 13.3% of all pts treated with BR and 18.6% with RHCOP (p=0.002). Autologous stem cell transplant (ASCT) was used in 26 (8.4%) POD 24 pts, among whom there was no difference in ASCT rate bt BR vs RCHOP (8.2% vs 8.8%, p=0.9). Conclusions: This retrospective analysis of real-world data showed clinical improvement in TTNT w BR compared to RCHOP in all FL patients. Among patients w G 3A FL, there was no significant improvement in TTNT with RCHOP. Rituximab mtn was associated w improved OS. Rate of transformation was higher in G 3A patients treated with RCHOP. There was low rate of ASCT in POD24 pts.

Extracellular (EC) Mucin5AC (MUC5AC) detection after neoadjuvant therapy (NAT) and the risk of recurrence in resected pancreatic ductal adenocarcinoma (R-PDA).

4171 Background: The clinical significance of MUC5AC glycoforms is unclear to date. Our earlier study showed the association of lower mature (MM) and immature (IM) MUC5AC in R-PDA with pathological treatment response and EC detection of MM (EC-MM). Here, we have the significance of EC-MM detection in the tissue of R-PDA post-NAT. Methods: Formalin-fixed paraffin-embedded tissue blocks of R-PDA from January 2010 to June 2021 were obtained from the Ohio State University biorepository. Using monoclonal antibodies for 45M1 (MM) and CLH2 (IM), immunohistochemistry was performed to study the cellular expression of MM and IM, and H-scores were calculated. Also, the presence or absence of EC expression of MM is noted. progression-free survival (PFS) and overall survival (OS) were the primary endpoints. Univariate and multivariate Cox regression models were used to model PFS and OS, and clinical and pathological variables were adjusted in the multivariate models. We performed univariate logistic regression analysis to examine the impact of MUC5AC expression on pathological features in resected specimens. Results: Among 100 R-PDA patients, 43 received NAT and 57 had upfront surgery (UpS)). In NAT-group (36-FOLFIRINOX, 2-FOLFOX, and 5-gemcitabine/nab-paclitaxel), EC-MM detection in the resected specimens was associated with worse PFS on multivariate analysis (MVA) (Hazard Ratio (HR) 0.2, 95 % CI: 0.059-0.766, p = 0.01). In the FOLFIRINOX-treated group, the EC-MM detection had a similar impact (HR 0.2, 95 % CI: 0.052-0.847, p = 0.02) on PFS. Cellular MM and IM expression were significant (p < 0.05) on MVA for OS in both groups, but the HRs were not clinically significant (NAT-group, MM - 0.9 & IM - 1.04, FOLFIRNIOX-group, MM - 0.8 & IM -1.02). Expression levels of cellular MM and IM in FOLFIRINOX group and MM alone in NAT-group significantly (p < 0.05) correlated with incomplete resection (R1/R2 vs R0) and perineural invasion. In the UpS group, MUC5AC glycoforms did not significantly correlate with survival or pathological features. Conclusions: Post NAT EC-MM detection and cellular MUC5AC (MM and IM) expression levels are associated with worse PFS and pathological features. Their role in the treatment of naïve patients is still unclear. MUC5AC can be a clinically useful predictive biomarker. Future research should focus examining its importance in advanced tumors and manipulating it for therapeutic purposes.

Survival outcome prediction of breast carcinomas on whole-slide histopathology images using deep learning.

1070 Background: Breast cancer is the most common cancer among women with 2 million new cases and 627,000 deaths in 2020. Early diagnosis and effective treatment are crucial for improved outcomes. Prognosis mainly depends on histopathological features, among them grade. As whole slide histopathology image (WSI) of tumor tissue contains a huge amount of hidden morphological features unexploited by pathologists, we investigate the potential of Artificial Intelligence (AI)-based analysis on WSI to predict prognosis in terms of 5-year overall survival in breast cancer patients. Methods: We used a novel deep neural network (DNN), DiaDeepBreastPRS, designed specifically for predicting a survival risk score in breast cancer patients based on H&E-stained whole slide images (WSI) of tumor tissues. The model incorporates two distinct viewpoints, one capturing cellular details and the other the tissue-level information. DiaDeepBreastPRS was trained and evaluated on a multi-centric discovery cohort of 1,027 patients (1,095 H&E WSI) from the TCGA-BRCA dataset using a cross-testing and cross-validation technique. It was evaluated on an external cohort, comprising 232 patients (247 H&E WSI). A statistical analysis with a multivariate Cox regression model was carried out on clinico-pathological data. AI scores and the concordance index (c-index) serves as the metric for assessing the performance. The predicted risk scores were used for effective risk stratification of breast carcinomas. Results: On the TCGA-BRCA dataset, the model achieved an average c-index of 67%, which reaches 78% by adding the pTNM stage and age at diagnosis. On the external dataset, the model achieved a c-index of 66% and 75% when including some histopathological prognosis factors (pTNM stage, age at diagnosis, HER2 and HR status and mitosis). The AI score was independently associated with the survival of breast cancer patients with the highest hazard ratio (HR 2.46, p<0.005). Furthermore, the model was able to significantly discriminate between the 2 groups of patients, with a good and a poor prognosis in terms of overall survival (p<0.005). Conclusions: In this study, we showcased that the algorithm was able to instantly extract prognostic morphological features from H&E whole slide images (WSI) and could be included in the pathology report. This could potentially enhance clinical decision-making, elevating the standard of care. Compared to commonly used molecular signatures, the AI algorithm enables a reduction in response time and cost savings. However, further investigations using additional independent cohorts are essential to consolidate the algorithm's performance and allow its generalizability, establish its superiority over existing prognostic markers, and provide insights into its interpretability.

Development of a multiparametric prognostic score based on PAM50, in patients with HR+/HER2- advanced breast cancer (aBC) treated with endocrine therapy (ET) plus cyclin dependent kinase 4 and 6 inhibitors (CDKi), within the CDK-PREDICT (SOLTI-1801) study and HCB cohort.

1061 Background: CDKi + ET combinations have demonstrated efficacy in improving progression-free survival (PFS) and overall survival (ribociclib) in HR+/HER2- aBC in the first-line setting. Intrinsic subtypes (IS) are prognostic and predict benefit from CDKi + ET. The multiparametric PAM50MET score has demonstrated prognostic value prior to the widespread use of CDKi in HR+/HER2- aBC patients. Here we developed a new prognostic score based on PAM50 assay from metastatic baseline biopsies prior to CDKi + ET within the CDK-PREDICT (SOLTI-1801) study and the Hospital Clinic de Barcelona (HCB) cohort. Methods: RNA from FFPE tumors was analyzed at the nCounter (Nanostring Technologies) using a 72 custom gene panel including the PAM50 genes. Intrinsic subtyping (luminal A, luminal B, HER2-enriched, basal-like, and normal-like) was performed using the research-based PAM50 IS predictor. The following parameters were used to develop the predictive score: PAM50 subtype assignments, PAM50 subtype expression signatures (PAM50 scores), expression of 72 individual genes included in the custom panel (PAM50 and no PAM50 genes) and 14 clinical variables. Parameters were assessed individually and all in combination. The most parsimonious model with the highest Harrel C-index (C-index) was selected. The prognostic value of the new model was assessed using multivariable Cox regression models (adjusting by endocrine sensitivity, visceral disease, and de novo metastatic disease) as a continuous variable and as a categorical variable. Results: Between May 2020 and May 2022, 185 patients (113 from CDK-PREDICT study and 72 from HCB cohort) diagnosed with HR+/HER2- aBC were evaluated. The combined model (CDK-SCORE) that achieved the highest C-index (table) included 4 clinical variables (ECOG, liver metastases, progesterone receptor expression and Ki-67), 6 genes as a continuous variable (AR, BRCA1, CCND1, CCNE1, PGR and FGFR4) and HER2-enriched and basal-like subtypes PAM50 scores. CDK-SCORE as a continuous variable was associated with PFS (aHR: 2.77, 95% CI 2.12 – 3.61). The median PFS of CDK-SCORE LOW patients was 32.9 months (95% CI 23.3 – NA) vs 16.3 months (95% CI 12.1 – 19.7) in CDK-SCORE HIGH patients (aHR: 2.76, 95% CI 1.74 – 4.37). CDK-SCORE was also categorised by tertiles and quartiles and mPFS and Kaplan-Meier curves will be presented. Conclusions: The combined CDK-SCORE demonstrates prognostic value in patients diagnosed with HR+/HER2- aBC and treated with ET+CDKi in the first-line setting. Future plans include conducting an external validation of the CDK-SCORE. [Table: see text]

Impact of breaks in therapy on survival for pediatric patients with relapsed or refractory solid tumors: A single-center study.

10021 Background: Relapsed or refractory (R/R) solid tumors remain a significant cause of mortality for children. While upfront therapies have been well-studied, the cumulative effects of multiple relapses and salvage regimens is unknown. We describe the disease trajectory in these patients, from the time of first R/R event (on-therapy progression/refractoriness or off-therapy relapse) to time of death or last follow-up. We also analyze whether breaks ( > 30 days) between diagnosis of events and initiation of subsequent therapies have an impact on survival. Methods: We reviewed data from electronic medical records for patients with primary, malignant R/R extracranial solid tumors treated at Texas Children’s Hospital between 2005-2023. Descriptive analysis was performed along with univariate chi-square, one-way ANOVA, and independent sample t-tests. Cox regression analysis was used to evaluate time from first event to death or last known follow-up for each variable described. Analysis was performed with SPSS v29. Results: We reviewed data for 466 patients with R/R solid tumors (female: 47%; median age: 8.66 years; median follow-up after first event: 12.9 months). Most common diagnoses were neuroblastoma (21.5%), rhabdomyosarcoma (19.1%), and osteosarcoma (16.1%). Patients had a median of 3 (interquartile range [IQR]: 1-4) R/R events with a median of 93 (IQR 52-200) days between each event. The most common post-event therapy regimens were intravenous chemotherapy alone (21.4%) or in combination with surgery and/or radiotherapy (19.9%). Deceased patients (n = 339) had a median of 291 (IQR: 127-549) days from first event to death. Increased time from first R/R event to death was associated with a break in disease-directed therapy; type of diagnosis; non-metastatic stage at original diagnosis; type of first event; type of first salvage regimen; and receiving phase 1 therapy. Of 438 patients who received ≥1 salvage therapy regimen, 116 (26.5%) took ≥1 break between an R/R event and subsequent therapy. Reasons for breaks varied relatively evenly between intentional/goal-concordant (e.g., to seek second opinion, to focus on quality of life), and unintentional (e.g., care coordination, care access issues). In a multivariable Cox regression model, taking a > 30 day break in disease-directed therapy was associated with prolonged survival, even after adjusting for number of events and significant covariates (p < 0.05). Conclusions: The prognosis for children with R/R solid tumors is poor. Most children experience multiple R/R events and therapy regimens without a break in therapy. Presence of a > 30 day break between an R/R event and subsequent therapy did not negatively impact survival, implying that time between therapy regimens can safely be offered to some patients to promote quality of life, exploration of goals of care, and thoughtful decision-making, without compromising survival.

Comparison of antitumor efficacy of first-line palbociclib, ribociclib, or abemaciclib in patients with HR+/HER2- aBC: Results of the multicenter, real-world, Italian study PALMARES-2.

1014 Background: The Cyclin Dependent Kinase 4/6 inhibitors (CDK4/6i) Palbociclib, Ribociclib and Abemaciclib in combination with Endocrine Therapy (ET) represent the standard-of-care, 1st line treatment for patients with Hormone Receptor-positive, Human Epidermal growth factor Receptor 2-negative, advanced Breast Cancer (HR+/HER2- aBC). So far, no large real-world studies have compared the efficacy of the three CDK4/6i in this clinical setting. Methods: The multicenter, population-based study PALMARES-2 evaluated the antitumor efficacy of 1st line Palbociclib, Ribociclib or Abemaciclib in combination with ET in consecutive HR+/HER2- aBC patients treated in 18 Italian cancer centers between 1st January 2016 and 1st September 2023. The primary study endpoint was real-world Progression-Free Survival (rwPFS), as defined as the time interval between ET plus CDK4/6i initiation and disease progression. Multivariate Cox regression model was used to adjust the association between individual CDK4/6i and rwPFS for clinically relevant variables. Results: With a data cut-off date of January 1st, 2024, we enrolled 1850 patients, 750 (40.6%), 676 (36.5%) and 424 (22.9%) of whom received Palbociclib, Ribociclib and Abemaciclib, respectively. Among 1226 (66.3%) patients with endocrine-sensitive disease, 1087 (89%) received concomitant Aromatase Inhibitors, whereas 441 (70.7%) out of 624 (33.7%) patients with endocrine-resistant disease received concomitant Fulvestrant. Patients treated with Abemaciclib were more likely to have endocrine-resistant disease, liver metastases, lobular tumor histology and lower PgR tumor expression, and less likely to have de novometastatic disease (p<0.001). Median rwPFS in the whole study cohort was 34.9 months (95% CI 32.0-37.4). Abemaciclib and Ribociclib were independently associated with better rwPFS when compared to Palbociclib (Table), whereas Abemaciclib and Ribociclib showed no significantly different efficacy (aHR 0.91, 95% CI 0.70-1.19; p=0.505). Other covariates independently associated with rwPFS are shown (Table). Conclusions: The real-world study PALMARES-2 revealed different antitumor efficacy of individual CDK4/6i in HR+/HER2- aBC patients. Longer follow-up is required to study if Palbociclib, Ribociclib and Abemaciclib are associated with different overall survival. [Table: see text]

High stress hyperglycemia ratio predicts adverse clinical outcome in patients with coronary three-vessel disease: a large-scale cohort study

BackgroundCoronary three-vessel disease (CTVD) accounts for one-third of the overall incidence of coronary artery disease, with heightened mortality rates compared to single-vessel lesions, including common trunk lesions. Dysregulated glucose metabolism exacerbates atherosclerosis and increases cardiovascular risk. The stress hyperglycemia ratio (SHR) is proposed as an indicator of glucose metabolism status but its association with cardiovascular outcomes in CTVD patients undergoing percutaneous coronary intervention (PCI) remains unclear.Methods10,532 CTVD patients undergoing PCI were consecutively enrolled. SHR was calculated using the formula: admission blood glucose (mmol/L)/[1.59×HbA1c (%)–2.59]. Patients were divided into two groups (SHR Low and SHR High) according to the optimal cutoff value of SHR. Multivariable Cox regression models were used to assess the relationship between SHR and long-term prognosis. The primary endpoint was cardiovascular (CV) events, composing of cardiac death and non-fatal myocardial infarction (MI).ResultsDuring the median follow-up time of 3 years, a total of 279 cases (2.6%) of CV events were recorded. Multivariable Cox analyses showed that high SHR was associated with a significantly higher risk of CV events [Hazard Ratio (HR) 1.99, 95% Confidence interval (CI) 1.58–2.52, P < 0.001). This association remained consistent in patients with (HR 1.50, 95% CI 1.08–2.10, P = 0.016) and without diabetes (HR 1.97, 95% CI 1.42–2.72, P < 0.001). Additionally, adding SHR to the base model of traditional risk factors led to a significant improvement in the C-index, net reclassification and integrated discrimination.ConclusionsSHR was a significant predictor for adverse CV outcomes in CTVD patients with or without diabetes, which suggested that it could aid in the risk stratification in this particular population regardless of glucose metabolism status.

Effectiveness against severe COVID-19 of a seasonal booster dose of bivalent (original/Omicron BA.4-5) mRNA vaccines in persons aged ≥60 years: Estimates over calendar time and by time since administration during prevalent circulation of different Omicron subvariants, Italy, 2022–2023

Evaluating how a COVID-19 seasonal vaccination program performed might help to plan future campaigns. This study aims to estimate the relative effectiveness (rVE) against severe COVID-19 of a seasonal booster dose over calendar time and by time since administration.We conducted a retrospective cohort analysis among 13,083,855 persons aged ≥60 years who were eligible to receive a seasonal booster at the start of the 2022–2023 vaccination campaign in Italy. We estimated rVE against severe COVID-19 (hospitalization or death) of a seasonal booster dose of bivalent (original/Omicron BA.4-5) mRNA vaccines by two-month calendar interval and at different times post-administration. We used multivariable Cox regression models, including vaccination as time-dependent exposure, to estimate adjusted hazard ratios (HR) and rVEs as [(1-HR)X100].The rVE of a seasonal booster decreased from 64.9% (95% CI: 59.8–69.4) in October-November 2022 to 22.0% (95% CI: 15.4–28.0) in April-May 2023, when the majority of vaccinated persons (67%) had received the booster at least 4–6 months earlier. During the epidemic phase with prevalent circulation of the Omicron BA.5 subvariant, rVE of a seasonal booster received ≤90 days earlier was 83.0% (95% CI: 79.1–86.1), compared to 37.4% (95% CI: 25.5–47.5) during prevalent circulation of the Omicron XBB subvariant. During the XBB epidemic phase, rVE was estimated at 15.8% (95% CI: 9.1–20.1) 181–369 days post-administration of the booster dose. In all the analyses we observed similar trends of rVE between persons aged 60–79 and those ≥80 years, although estimates were somewhat lower for the oldest group.A seasonal booster dose received during the vaccination campaign provided additional protection against severe COVID-19 up to April-May 2023, after which the incidence of severe COVID-19 was much reduced. The results also suggest that the Omicron XBB subvariant might have partly escaped the immunity provided by the seasonal booster targeting the original and Omicron BA.4-5 strains of SARS-CoV-2.

Prognostic significance of real-world (RW) recurrence patterns after resection or ablation of early-stage hepatocellular carcinoma (HCC): Insights from the US Medicare population.

e16241 Background: The prognosis after curative-intent resection or ablation of early-stage HCC remains poor, largely due to high rates of recurrence. We aimed to characterize real-world recurrence patterns and determine whether the timing of recurrence was associated with overall survival (OS). Methods: Patients (pts) with early-stage HCC who underwent resection and/or ablation were identified from the SEER-Medicare database (2010-2019). Pts who received liver transplant were excluded. A proxy for recurrence was defined as the receipt of new HCC treatments (surgery, locoregional therapy, systemic therapy) or a secondary malignant neoplasm claim (excluding liver). Pseudo-recurrence dates were assigned to pts without recurrence to match the distribution of time between initial resection/ablation and recurrence in the recurrence cohort, to avoid immortal time bias. Early vs late recurrence was defined as occurring within 1 year vs after 1 year of initial resection/ablation. Primary outcome was OS as measured both from the time of recurrence/pseudo-recurrence and initial resection/ablation. Results: 1,146 eligible pts with early-stage HCC were identified. 63% were male, mean age was 74 yrs, and initial therapy was resection (39%), ablation (50%), or both (11%). 736 (64%) experienced recurrence. Among them, 53% received ablation, 37% surgery, and 11% received both. Median OS from recurrence or pseudo-recurrence was 22 months (mo) vs 67 mo (p &lt; 0.001) for pts with vs without recurrence, respectively. When measured from resection/ablation date, median OS was 39 mo vs 83 mo (p &lt; 0.001). After adjusting for relevant demographic and clinico-pathologic factors, including type of index therapy, in a multivariable Cox regression model, recurrence remained significantly associated with an increased risk of death (hazard ratio [HR]: 2.24; 95% CI: 1.85-2.71; p &lt; 0.001). Among pts who recurred, 380 (52%) had early recurrence and 356 (48%) had late recurrence. Median OS from recurrence was 17 mo vs 30 mo (p &lt; 0.001) for early vs late recurrence, respectively, and 23 mo vs 63 mo when measured from resection/ablation The adjusted HR for the risk of death associated with early recurrence was 1.39 (95% CI: 1.14-1.68; p &lt; 0.001) compared to late recurrence. Other significant prognostic factors associated with reduced OS included larger tumor size ( &gt; = 5cm), greater extent of disease, and poor differentiation (grade III and above). Conclusions: In this study among US Medicare pts, early recurrence within one year after curative-intent treatment of early-stage HCC was significantly associated with decreased OS compared to pts who recur later or without recurrence. This highlights the need for developing and utilizing effective adjuvant therapy regimens to delay or avoid recurrence.

Association of combination tumor texture and vessel tortuosity with progression free survival across PD-L1 subgroups in durvalumab treated non-small cell lung cancer (NSCLC): Blinded validation results from CP1108.

8600 Background: Immunotherapy (IO) has shown a durable response with minimal toxicity in the treatment of metastatic NSCLC. However, only 30-50% of these patients (pts) actually respond. Despite initial optimism surrounding PD-L1 expression as a potential biomarker, IO provides benefit across PD-L1 low (&lt;25%) and high (≥25%) status. This highlights a critical need for more effective biomarkers to guide treatment decisions. Here, we present blinded validation results of a CT-based biomarker of change in quantitative vessel tortuosity (1) (Δ-QVT) and texture radiomics (2) (Δ-Rad) between baseline (B) and 6 weeks post-treatment (TP1) for predicting response, overall survival (OS) and progression free survival (PFS) in CP1108 (NCT01693562). Methods: CT scans at B and TP1 were retrospectively analyzed from two studies (a) a multi-center training set (Str, N=110) of metastatic NSCLC pts treated with first line IO and (b) a blinded NSCLC cohort of N = 151 pts treated with Durvalumab in CP1108 study (Sv). In Sv, N=75 pts were PD-L1 high (Sv+) whereas N=65 pts were PD-L1 low (Sv-). An in-house MATLAB based algorithm was used to extract intra-tumoral, peri-tumoral texture and QVT features from up to two largest measurable tumors in each CT. A classifier (MCombo) was trained on Str to predict best overall response defined as per the RECIST v1.1 criteria using combination of Δ-QVT and Δ-Rad features. Area under the receiver operating characteristic (AUC) was used to evaluate MCombo against RECIST response whereas univariable and multivariable Cox regression models with log-rank test, hazard ratio (HR) with confidence interval (CI) and concordance index (C-index) were used to study the association of MCombo with OS and PFS in Sv, Sv+, and Sv-. Results: MCombo predicted objective response with an AUC of 0.78 in Sv and was statistically significantly associated with PFS (Table) in Sv, Sv+, and Sv-. MCombo was associated with OS in Sv, Sv+, but not in Sv-. Multivariable analysis with PD-L1 status, histological subtype, ECOG status, liver metastasis, line of therapy and sex showed independent association of MCombo across OS and PFS. Conclusions: Our results indicate that combination of radiomics and vessel tortuosity biomarker from CT is independently prognostic of response and PFS across PD-L1 levels. Prospective validation of this biomarker is needed. 1. Alilou et al. Sci Adv 2022. 2. Khorrami et al. Cancer Immunol Res 2020. [Table: see text]

Association between social determinants of health and time to treatment for newly diagnosed multiple myeloma.

1583 Background: Delayed access to costly anti-myeloma agents due to insurance precertification process or out-of-pocket costs results in treatment initiation with less-costly corticosteroids in some patients with newly diagnosed multiple myeloma (NDMM). Less is known about the scope of the delay and the factors associated with it. Methods: This retrospective cohort study used Taussig Cancer Center’s Myeloma Patient Registry to identify adults with NDMM between January 1, 2017 and December 31, 2021. Electronic health records data captured (1) time from initial diagnosis to receipt of FDA-approved anti-myeloma agent including vs. excluding corticosteroids, (2) time to initial prescription fill for FDA-approved anti-myeloma drug excluding corticosteroids, and (3) association of race, sex, primary insurance type, area deprivation index (ADI), urbanicity, and time to initial fill for anti-myeloma drug excluding corticosteroids. The latter was examined via cumulative incidence at 30 days since initial diagnosis using the log-rank test and multivariable Cox regression model. Results: We identified 723 patients with a mean age at diagnosis of 67 years ±11; 55% were male, 77% White, 22% Black, and 1% other races, covered by private insurance (36%), traditional Medicare (28%), Medicare Advantage (25%), Medicaid (8.4%), and self-pay/other (1.8%); and median follow-up of 784 days (interquartile range [IQR], 415, 1313). Over a third of patients (37%) resided in an area in the most disadvantaged quartile of the ADI; 83% resided in a metropolitan area. Median time to receipt of anti-myeloma treatment was 10 days (IQR, 2, 24) including corticosteroids vs. 18 days (IQR, 8, 37) excluding corticosteroids. Median time to initial prescription fill for an anti-myeloma drug excluding corticosteroids was 31 days (IQR, 16, 75). The cumulative incidence of prescription fill for anti-myeloma drug excluding corticosteroids at 30 days was 40% (95% CI, 36%, 44%) in White individuals, 28% (95% CI, 21%, 36%) in Black patients, and 17% (95% CI, 0.42%, 64%) in other races (P=.009) [adjusted HR for Black patients (v. White), 0.63, 95% CI, 0.44, 0.90]; 31% (95% CI, 25%, 38%) in patients with traditional Medicare, 37% (95% CI, 30%, 44%) with Medicare Advantage, 44% (95% CI, 38%, 50%) with Private Insurance, 34% (95% CI, 23%, 48%) with Medicaid, and 31% (95% CI, 9.1%, 61%) self-paying/with other insurance (P=.07) [adjusted HR for private insurance (v. traditional Medicare), 1.41, 95% CI, 1.04, 1.93]; 40% (95% CI, 36%, 44%) treated in Taussig Cancer Center and 25% (95% CI, 17%, 34%) in regional hospitals (P=.003) [adjusted HR for regional hospitals, 0.59, 95% CI, 0.40, 0.89]. Conclusions: Disparities in time to prescription fill for anti-myeloma drugs exist based on socio-demographic factors. Future research should examine the role of precertification criteria and out-of-pocket costs in delays with recommended care.

TTF-1 negativity as a biomarker of outcomes to immunotherapy, chemoimmunotherapy, and KRAS G12C inhibitors in lung adenocarcinoma.

8608 Background: Thyroid transcription factor 1 (TTF-1) expression is routinely assessed in lung adenocarcinomas (LUAD) and is negative (TTF-1Neg) in 15-20% of cases. Whether these tumors differ from LUAD expressing TTF-1 (TTF-1Pos) in terms of biologic features and outcomes is unclear. Methods: Patients (pts) with LUAD and available TTF-1 expression by IHC at five institutions were included. Clinicopathologic, genomic, and outcomes data were compared according to TTF-1 expression. Results: Among 3,315 pts with LUAD, TTF-1 was negative in 15% (N=499). Compared to TTF-1Pos (N=2,816), pts with TTF-1Neg LUAD were more likely male (44% vs 38%, P=0.01), smokers (82% vs 74%, P&lt;0.001), with stage IV disease at diagnosis (76% vs 71%, P&lt;0.001), and lower median PD-L1 tumor proportion score (TPS, 1% vs 10%, P&lt;0.001). At stage IV diagnosis, TTF-1Neg cases less commonly had brain metastasis than TTF-1Pos cases (35% vs 44%, P=0.04), but more frequently had bone metastasis (53% vs 45%, P=0.03). TTF-1Neg LUAD was enriched for KRAS, STK11, KEAP1, SMARCA4, and CDKN2A mutations (q&lt;0.05), while TTF-1Pos LUAD was enriched for EGFR and MET mutations (q&lt;0.05). Compared to TTF-1Pos LUAD, TTF-1Neg LUAD had higher prevalence of KRAS G12D (12% vs 4%, P&lt;0.001) and G12V mutations (11% vs 7%, P=0.001), but similar frequency of G12C mutations (14% vs 15%, P=0.82). The association between TTF-1 and treatment outcomes was next investigated. Pts with stage III unresectable TTF-1Neg LUAD who received durvalumab after chemo-radiation (N=20), compared to TTF-1Pos cases (N=174), had shorter median progression-free survival (mPFS, 7.6 vs 24.7 months, P=0.01) and overall survival (mOS, 22.7 months vs not reached, P=0.01). Among pts with stage IV LUAD, TTF-1Neg cases treated with immune checkpoint inhibitors (N=237), compared to TTF-1Pos (N=1,161), had worse objective response rate (ORR, 17% vs 28%, P&lt;0.001), mPFS (2.5 vs 4.7 months, P&lt;0.001) and mOS (9.8 vs 20.9 months, P&lt;0.001). Similarly, pts with TTF-1Neg LUAD (N=297) treated with chemoimmunotherapy, compared to TTF-1Pos cases (N=920), had worse ORR (27% vs 42%, P&lt;0.001), mPFS (4.6 vs 8.1 months, P&lt;0.001) and mOS (11.2 vs 23.4 months, P&lt;0.001). TTF-1 negativity retained its association with worse outcomes after adjusting for treatment line, ECOG performance status, smoking status, PD-L1 TPS, STK11/ KEAP1 mutation status in multivariable cox regression models. Lastly, pts with KRAS G12C mutated TTF-1Neg LUAD treated with KRAS inhibitors (N=25), compared to TTF-1Pos (N=138), had worse ORR (12% vs 35%, P=0.03), mPFS (2.7 vs 5.9 months, P&lt;0.001), and mOS (4.4 vs 12.1 months, P&lt;0.001). After excluding mucinous LUAD cases, which are more commonly TTF-1Neg, similar results were observed. Conclusions: TTF-1 negativity identifies a unique clinicopathologic and genomic subset of LUAD with worse outcomes to diverse systemic therapies.

Use of chemotherapy and loco-regional therapy for stage IA triple-negative breast cancer and their association with oncologic outcomes: A cancer registry study.

524 Background: We aimed to evaluate the role of adjuvant chemotherapy and loco-regional therapy for stage IA (pT1, pN0) triple-negative breast cancer (TNBC) in a real-world setting. Methods: We identified patients with pT1, pN0 TNBC diagnosed between 2009 and 2021 within the clinical cancer registry of Baden-Wuerttemberg, Germany. Overall survival (OS) was assessed using Kaplan-Meier statistics and multivariate Cox regression models (adjusted for age, use of chemotherapy, local therapy (breast conserving therapy [breast conserving surgery + radiotherapy] vs. mastectomy), and tumor histologic subtype). Results: A total of 1191 patients with a median follow-up of 43.8 months were identified: 0.9% (11 of 1191) with pT1mi stage, 10.0% (119 of 1191) with pT1a, 24.0% (286 of 1191) with pT1b, and 65.0% (775 of 1191) with pT1c. Use of adjuvant chemotherapy was 9.1% (1 of 11) in pT1mi stage, 41.2% (49 of 119) in pT1a, 66.8% (191 of 286) in pT1b, and 69.8% (541 of 775) in pT1c. Multivariate Cox regression analysis revealed no significant influence for the use of adjuvant chemotherapy on OS in pT1a and pT1b patients (for pT1a HR 1.27, 95% CI 0.19 to 8.50, for pT1b HR 1.07, 95% CI 0.50 to 2.30) but for patient age in the pT1b group (HR 4.71, 95% CI 2.26 to 9.8, for age ≥70 vs. &lt;70 years). For pT1c patients, there was significant influence for the use of adjuvant chemotherapy on OS (HR 0.55, 95% CI 0.38 to 0.8), for local therapy (HR 0.41, 95% 0.25 to 0.69, for breast conserving therapy vs. mastectomy only), and patient age (HR 2.33, 95% CI 1.61 to 3.39 for age ≥70 vs. &lt;70 years). Notably, for pT1c patients aged &lt;70years, local therapy (HR 0.32, 95% CI 0.15-0.69), in favor of breast conserving therapy) but not use of chemotherapy (HR 1.11, 95% CI 0.56-2.19) was associated with OS. Conclusions: This data suggests that OS in stage IA TNBC is strongly influenced by local therapy and patient age, but not use of chemotherapy. Larger studies with longer-term follow-up are welcomed to fully inform this discussion.

Global retrospective study comparing consolidation ALK tyrosine kinase inhibitors (TKI) to durvalumab (durva) or observation (obs) after chemoradiation (CRT) in unresectable locally-advanced ALK+ non-small cell lung cancer (NSCLC).

8013 Background: For patients (pts) with unresectable locally-advanced NSCLC, standard of care involves durva consolidation after concurrent CRT, although its benefit in ALK+ tumors is unclear. The ALINA trial demonstrated adjuvant efficacy of alectinib in resected ALK+ NSCLC, but the optimal consolidation strategy for unresectable locally-advanced ALK+ NSCLC remains elusive. Methods: This multi-institutional international retrospective analysis included pts with stage III unresectable ALK+ NSCLC who received an ALK TKI, durva, or obs alone after concurrent CRT between 2015-2022. Baseline characteristics of age, sex, smoking history, and PD-L1 status were collected. Clinical outcomes including real-world progression-free survival (rw-PFS), overall survival (OS), and treatment-related adverse events (trAE) as defined using CTCAE 5.0 were assessed, and multivariate cox regression models were performed. Results: Sixty-four pts across 16 institutions were included. The median age was 57 (IQR 49-66) and 61% were females. The majority had adenocarcinoma (97%) and had never smoked (58%). 15 received ALK TKI (10 alectinib, 3 crizotinib, 1 brigatinib, 1 lorlatinib), 30 received durva, and 19 received obs alone. There was no significant difference in stage of cancer (IIIA, B, or C) or PD-L1 status among groups. After adjusting for stage and age at CRT initiation, median rw-PFS was significantly longer for ALK TKI (rw-PFS not reached [NR], 95% CI 22.7-NR) vs durva (11.3 months (mo), 95% CI 9.2-18.5, p= 0.005, HR = 0.12) or obs (7.4 mo, 95% CI 3.4-12.5, p &lt; 0.0001). Two (13%) pts progressed on ALK TKI. 25 pts (83%) progressed on durva leading to treatment with ALK TKI in 23, and 18 (95%) progressed while under obs leading to treatment with ALK TKI in 15. 3-year OS was 100% for ALK TKI vs 90.5% for durva vs 63.5% for obs. Median OS was NR (95% CI NR-NR) for both ALK TKI and durva vs 70.6 mo (24.9-NR) in the obs group ( p= 0.03 for both ALK TKI and durva compared to obs). Median duration of therapy was 24.7 mo with ALK TKI and 6.5 mo with durva. Grade ≥3 trAE occurred in 27%, 7%, 6% of pts treated with ALK TKI (1 fatigue, 1 diarrhea, 1 hyperbilirubinemia, 1 pneumonitis), durva (1 fatigue, 1 neutropenia), and obs (1 neutropenia) respectively. Treatment discontinuation due to toxicity occurred in 4 (27%) with ALK TKI and in 3 (10%) with durva. Conclusions: In this retrospective study of stage III ALK+ NSCLC, consolidation ALK TKI demonstrated clinically meaningful improvement in PFS and OS over durva and obs, while showing a slightly higher rate of trAE over dura and obs. Furthermore, we show a high rate of progression following CRT alone in ALK+ NSCLC. These findings underscore the need for prospective molecularly driven trials to determine the optimal consolidation therapy for unresectable ALK+ NSCLC.

The association of hormone receptor expression level and clinical efficacy of CDK4/6 inhibitor in hormone receptor-positive, HER2-negative metastatic breast cancer.

e13069 Background: In patients with hormone receptor-positive, HER2-negative metastatic breast cancer (HR+ HER2- MBC), the correlation between the level of expression of HR and the clinical efficacy of CDK4/6 inhibitors remains undefined. Methods: We identified consecutive patients with HR+ HER2- MBC starting a CDK4/6 inhibitor (palbociclib, ribociclib or abemaciclib) plus either an aromatase inhibitor or fulvestrant between Jan 2018 - Oct 2023 from an institutional cancer registry. Expression level of HR was stratified as high (H) (Allred score 7-8), moderate (M) (Allred score 5-6) or low (L) (Allred score 3-4) for both estrogen (ER) and progesterone receptors (PR). Patients were evaluated from start of treatment to December 2023, last visit or death, whichever came earlier. Median progression-free survival (mPFS) and overall survival (mOS) were endpoints. We evaluated the association of HR expression levels with mPFS and mOS using log-rank test and multivariate Cox regression modelling. Results: We examined data from 436 women with HR+ HER2- MBC. Median age at treatment was 60.5 yr. Majority of patients received the treatment in the first-line setting (72.5%), with palbociclib (54.1%), ribociclib (31.9%) and abemaciclib (14.0%) as the CDK4/6 inhibitor. In the population that received first-line treatment, ER expression levels were H (83.5%), M (11.4%) and L (5.1%) respectively. ER expression levels were strongly associated with mPFS and mOS. For ER H, M and L expressions, mPFS were 29.4m, 9.2m and 2.9m ( p&lt;0.001); mOS were 52.1m, 40.6m and 11.6m ( p&lt;0.001), respectively. In multivariate analysis, ER expression levels remain strongly predictive for both mPFS and mOS after adjustment for clinicopathological risk factors (ER L vs H, mPFS HR 10.6, p&lt;0.001; mOS HR 4.66, p&lt;0.001; ER M vs H, mPFS HR 4.90, p&lt;0.001; mOS HR 2.06, p=0.05). Findings were similar for the second line and beyond population. In the ER-high expression cohort, the PR expression levels also correlate strongly with treatment efficacy and survival, with mPFS of 40.8m, 24.5m, 24.0m, 14.1m ( p&lt;0.001) and mOS of NR, 53.3m, 48.1m, 26.9m ( p&lt;0.001) respectively for PR H, M, L and zero expressions. Conclusions: In this real-world cohort of patients with HR+ HER2- MBC, we observed that the semi-quantitative HR expression level subgrouping is a strong predictor for the efficacy of treatment with CDK4/6 inhibitors. In contrast to patients with high HR expression levels, the small population with ER-low expression derived very limited benefits from the standard endocrine therapy with CDK4/6 inhibitors. Our findings provided a practical tool to guide the selection of the appropriate systemic therapy for individual patients.

Carpal Tunnel Syndrome and Trigger Finger May Be an Early Symptom of Preclinic Type 2 Diabetes.

Type 2 diabetes (T2D) is a major risk factor for carpal tunnel syndrome (CTS) and trigger finger (TF), but less is known regarding the risk of developing T2D after being diagnosed with CTS or TF. CTS and TF could be early signs of preclinical T2D, and early detection of T2D is crucial to prevent complications and morbidity. Therefore, we investigate the association between CTS/TF and T2D in an adult population without previous T2D using big data registers in Sweden. Data were collected by crosslinking five nationwide Swedish registers. Individuals aged 40-85 years on December 31, 2010, without prior overt diabetes, were included (n = 3,948,517) and followed up from baseline (ie, a diagnosis of CTS or TF) or January 1, 2011, for controls, until a diagnosis of T2D, prescription of oral antidiabetics or insulin, or end of follow-up four years after baseline. Multivariate Cox regression models were created to calculate hazard ratios for T2D. In total, 37,346 (0.95%) patients were diagnosed with CTS, whereof 1329 (3.46%) developed T2D. There were 17,432 (0.44%) patients who developed TF, whereof 639 (3.67%) developed T2D. Among the controls, 2.73% developed T2D. Compared with controls, there was an increased risk of developing T2D after being diagnosed with either CTS (HR 1.35; 95% confidence interval 1.28-1.43) or TF (HR 1.21; 95% confidence interval 1.12-1.31). Compared with controls, a diagnosis of CTS or TF was associated with 35% and 21% higher risk for later T2D, respectively, which might indicate the existence of undetected T2D in this population.

The efficacy of cumulative cisplatin dose on survival in patients with locally advanced cervical cancer treated with definitive chemoradiotherapy: A multicenter study by the Turkish Oncology Group (TOG).

e17534 Background: This multicentric study aimed to investigate the impact of cumulative cisplatin dose on clinical outcomes in locally advanced cervical cancer patients undergoing definitive chemoradiotherapy (CRT). Methods: A retrospective analysis was conducted on 654 patients with stage IB3-IVA disease treated with definitive CRT. Radiotherapy (RT) was applied as external beam pelvic with or without para-aortic RT and brachytherapy. Concomitant chemotherapy was in the form of weekly or 3-weekly cisplatin. Data on demographics, treatment protocols, cumulative cisplatin dose, adverse effects, and survival outcomes were collected. Statistical analyses, including univariate and multivariate Cox regression models, were employed to assess factors influencing progression-free survival (PFS) and overall survival (OS). Results: The median cumulative cisplatin dose was 210 mg (range, 40-320 mg), and it was ≥200 mg in 503 (76.9%) patients. Median follow-up was 35 months. The 5-year PFS and OS rates were 66.9% and 77.1%, respectively. Multivariate analysis (MVA) identified poor performance status, non-squamous cell histology, presence of lymph node metastasis and hemoglobin &lt;10 g/dl before CRT were poor prognostic factors for both PFS and OS in the whole group. When stage III cases were evaluated separately, the cumulative cisplatin dose below 200 mg was also found to be a significant poor prognostic factor in OS (HR 1.79, 95% CI, 1.1 to 3.0, p=0.031). Conclusions: The study revealed that a cumulative cisplatin dose exceeding 200 mg, particularly in patients with lymph node metastasis, significantly improved OS. Factors such as anemia, toxicity-related challenges, and comorbidities were identified as critical considerations in treatment planning. These findings emphasize the delicate balance between maximizing therapeutic efficacy and managing toxicity, guiding personalized treatment approaches for locally advanced cervical cancer. [Table: see text]

Overall survival analysis of first line CDK4/6 inhibitors in a large real-world cohort of patients with hormone receptor-positive (HR+)/HER2-negative metastatic breast cancer.

e13095 Background: Multiple clinical trials have demonstrated the efficacy of the commercially available CDK4/6 inhibitors (CDK4/6i) combined with endocrine therapy (ET) as upfront therapy in patients (pts) with (HR+)/HER2-negative (HER2-) metastatic breast cancer (MBC). Recently, differences in overall survival (OS) benefit in MBC and invasive disease-free survival benefit in early-stage BC suggest greater potency of abemaciclib and ribociclib compared to palbociclib. Here we report a comprehensive OS analysis of a cohort of pts with HR+/HER2- MBC treated with ribociclib, palbociclib or abemaciclib and ET in 1st line setting. Methods: We retrieved medical records of pts with HR+/HER2- MBC receiving CDK4/6i as 1st line therapy between December 2013 and October 2023 at Memorial Sloan Kettering Cancer Center. We investigated the impact of clinico-pathological characteristics and treatment type as categorical variables on OS using uni- and multivariate Cox (proportional hazard) regression model. Any p-value (p) &lt;0.05 was deemed to be statistically significant. Results: 854 pts were included in the analysis, 838 females (98%) and 16 males at birth. The median age was 56 (IQR: 48-65). Overall, 722 (84%) pts received palbociclib, 88 (11%) abemaciclib, and 44 (5%) ribociclib. Most pts received aromatase inhibitor (AI) as ET backbone (n=634, 74%); 214 received fulvestrant (35%), and 6 received tamoxifen (1%). After a median follow-up of 3.14 years (range: 1.7-4.6), the median OS of the full cohort was 54.7 months (95%CI: 50-63.1). Clinico-pathological characteristics that resulted independently associated with worse OS were: male gender at birth (HR:2, 95%CI:1.4-4), disease-free interval ≤ 2 years (HR:1.64, 95%CI:1.11-2.44), ER-low status (HR: 2.27, 95%CI:1.42-3.63), high grade(HR: 1.28, 95%CI:1.01-1.62), liver involvement (HR: 1.48, 95%CI:1.08-2.03), and number of disease sites &gt;3 (HR:1.35, 95%CI:1.03-1.78). HER2-low status was associated with better OS (HR: 0.75, 95%CI:0.59-0.96). In terms of type of treatment, there was a trend toward improved OS in patients receiving AI rather than fulvestrant or tamoxifen (HR:0.76, 95%: 0.58-1), while the type of CDK4/6i did not affect the outcome. Conclusions: Clinical parameters of early relapse, high disease burden, and intrinsic tumor aggressiveness seem to be the main drivers of poor prognosis in patients with HR+/HER2- MBC in 1st line setting. In this analysis, type of CDK4/6i did not have a significant impact on OS, although pts treated with abemaciclib and ribociclib were underrepresented in the cohort due to the earlier and prevalent use of palbociclib as the first-in-class FDA-approved CDK4/6i. These data are reassuring about the use of palbociclib suggesting that a switch of CKD4/6i is not needed in pts with ongoing treatment and demonstrated benefit.