Subpopulation Treatment Effect Pattern Plot Research Articles

Abstract CT099: The benefit of abemaciclib in prognostic subgroups: An update to the pooled analysis of MONARCH 2 and 3

Abstract Background: A recent exploratory subgroup analysis pooling data from over 1000 patients (pts) in MONARCH 2 (final analysis) and MONARCH 3 (interim analysis) demonstrated that pts with clinical characteristics that predict for worse outcome (e.g. liver metastases, progesterone receptor [PgR]-negative status, high-grade tumors) received the largest benefit from the addition of abemaciclib to endocrine therapy (ET). There is also interest in determining the extent to which abemaciclib improves progression-free survival (PFS) in pts with favorable prognostic factors (e.g. bone-only disease, long treatment-free interval [TFI]). Here we present an update of the subgroup analyses of the MONARCH 3 study using data from the preplanned final PFS analysis. Methods: Enrollment criteria and study designs were previously described (Sledge et al. J Clin Oncol. 2017; Goetz et al. J Clin Oncol. 2017). An exploratory pooled analysis was conducted using data from MONARCH 2 (NCT02107703) and the MONARCH 3 (NCT02246621) preplanned final PFS analysis to identify significant prognostic factors regardless of therapy received. Using the updated MONARCH 3 data, PFS and objective response rate (ORR) (in pts with measurable disease) were examined within the prognostic subgroups that had been previously identified as being statistically significant. Additionally, subpopulation treatment effect pattern plot (STEPP) methodology was used to evaluate the association between TFI following adjuvant ET and outcomes of NSAI alone or with abemaciclib. Results: Assessment of clinical factors from MONARCH 2 and MONARCH 3 (final PFS data) confirmed that the following factors remained significantly prognostic regardless of treatment received: bone-only disease, liver metastases, tumor grade, PgR status, and ECOG performance status. In MONARCH 3, prognosis was poor in pts with liver metastases, PgR-negative tumors, and high-grade tumors. While all subpopulations benefited from the addition of abemaciclib to ET regardless of prognosis, pts with these three poor prognostic factors received substantial benefit from abemaciclib, with PFS hazard ratios (HR) ranging from 0.4 to 0.5 and with differential improvement in ORR typically >30%. Furthermore, STEPP analysis of TFI (on the MONARCH 3 pts who had received adjuvant ET) demonstrated that pts who had a shorter TFI had a worse prognosis and received relatively greater benefit from abemaciclib plus NSAI than did pts with a longer TFI. Patients with bone-only disease or a long TFI also received benefit from abemaciclib, but to a relatively lesser extent (HR ranging from approximately 0.6 to 0.8). Conclusions: This exploratory analysis of the preplanned final PFS data from MONARCH 3 confirms previous findings that pts with poor prognostic factors may receive relatively greater benefit from the addition of abemaciclib to ET. Citation Format: Joyce O'Shaughnessy, Matthew P. Goetz, George W. Sledge, Miguel Martin, Yong Lin, Tammy Forrester, Ian C. Smith, Angelo Di Leo, Stephen Johnston. The benefit of abemaciclib in prognostic subgroups: An update to the pooled analysis of MONARCH 2 and 3 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr CT099.

Absolute improvements in freedom from distant recurrence with adjuvant endocrine therapies for premenopausal women with hormone receptor-positive (HR+) HER2-negative breast cancer (BC): Results from TEXT and SOFT.

503 Background: The TEXT and SOFT trials randomly assigned premenopausal women with HR+ BC to exemestane plus ovarian function suppression (E+OFS), tamoxifen+OFS (T+OFS) and T alone. We previously examined absolute treatment effects on any BC recurrence across a continuum of recurrence risk to individualize endocrine therapy decision making. After 8.5 yrs median follow-up we now consider effects on freedom from distant recurrence. Methods: The TEXT and SOFT HR+/HER2- analysis population included 4891 pts and was stratified by chemotherapy use. The endpoint was distant recurrence-free interval (DRFI; time from randomization to first appearance of DR). For each pt, a previously defined continuous, composite recurrence risk index (CRI) was calculated from a Cox model incorporating age, nodal status, tumor size and grade, and ER, PgR and Ki-67 expression levels. Subpopulation Treatment Effect Pattern Plot (STEPP) methodology estimated 8yr absolute treatment effects for subpopulations defined by CRI values. Results: Overall 8yr DRFI was 91% (433 DRs) and ranged from ~100% to 63% across lowest CRI of 0.2 to highest CRI of 3.4. TEXT pts (n = 2267) had median [IQR] CRI of 1.7 [1.2-2.3]. Overall 8yr DRFI was 92% (194 DRs) and absolute benefit of E+OFS vs T+OFS was 3%; benefit ranged from 0% at lowest CRI (both therapies having 100% 8yr DRFI) to 15% at highest CRI. SOFT pts who remained premenopausal after chemo (n = 1271) had median CRI of 2.1 [1.5-2.7]. Overall 8yr DRFI was 82% (216 DRs) and absolute benefit of E+OFS vs T was 5%; benefit ranged from 2% to 10% across CRI. For T+OFS vs T the absolute benefit ranged 0% to 5%. The SOFT no-chemo cohort (n = 1353) had median CRI of 1.1 [0.3-1.4]. Overall 8yr DRFI was 99% (23 DRs), and across CRI exceeded 97% with all three endocrine therapies. Conclusions: Premenopausal pts with HR+/HER2- BC and high recurrence risk, as defined by clinicopathological characteristics, may experience 10-15% absolute improvement in 8yr DRFI with E+OFS vs T+OFS or T alone. Potential benefit of escalating endocrine therapy vs T alone is minimal for those at low risk, and may be 4-5% for pts at intermediate risk. Clinical trial information: NCT00066690, NCT00066703.

Competing nomograms help in the selection of elderly patients with colon cancer for adjuvant chemotherapy.

The extent to which ≥ 70year patients with colon cancer benefit from adjuvant chemotherapy in the presence of competing risks remains controversial. 18,937 patients ≥ 70years with high-risk stage II and stage III colon cancer were retrospectively reviewed from SEER database. Propensity score matching (PSM) was used to adjust for potential baseline confounding. The nomograms were developed based on the competing model to describe the individual probability of colon cancer-specific death (CCSD) and non-CCSD. The subpopulation treatment-effect pattern plot (STEPP) was used to estimate the treatment-effect heterogeneity. In the high-risk stage II subgroup, compared to the non-recipients, the hazard ratios (HR) of overall mortality for recipients were 0.83 (P = 0.001). The subdistribution hazard ratio (SHR) of CCSD for receipts was 1.22 (P = 0.021). The SHR of non-CCSD was 0.63 (P < 0.001). In the stage III subgroup, compared to non-recipients, the HR of the overall mortality for the recipients was 0.62 (P < 0.001). The SHR of CCSD was 0.77 (P < 0.001). The SHR of non-CCSD was 0.58 (P < 0.001). The chemotherapy efficacy differed significantly by risk score of non-CCSD (non-CCSD-RS) (P < 0.001). Recipients with high non-CCSD-RS had a rate of CCSD comparative to that of non-recipients (SHR 0.90, P = 0.150) in the stage III subgroup. A survival analysis based on the overall mortality did not correctly interpret the effect of chemotherapy. Adjuvant chemotherapy did not provide an additional benefit to patients with high-risk stage II or patients with stage III at high risk of non-cancer death.

Abstract GS6-02: The benefit of abemaciclib in prognostic subgroups: An exploratory analysis of combined data from the MONARCH 2 and 3 studies

Abstract Background: Abemaciclib is an orally administered, selective inhibitor of cyclin-dependent kinases 4 &amp; 6 that is dosed on a twice daily continuous schedule. Abemaciclib has demonstrated clinical efficacy with a generally tolerable safety profile in patients with hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2-) advanced breast cancer in combination with fulvestrant in MONARCH 2 (NCT02107703) and in combination with non-steroidal aromatase inhibitors (NSAI) in MONARCH 3 (NCT02246621). These analyses were conducted to evaluate if patient and disease characteristics may better inform in whom and when abemaciclib should be initiated to define optimal treatment strategies. Methods: MONARCH 2 and 3 enrolled patients with HR+, HER2- advanced breast cancer. In MONARCH 2, patients whose disease had progressed while receiving endocrine therapy were treated with abemaciclib/placebo plus fulvestrant. In MONARCH 3, patients were treated with abemaciclib/placebo plus NSAI as initial therapy for advanced disease. An exploratory pooled analysis of the two studies was performed to determine significant prognostic factors. Efficacy results (progression-free survival [PFS] and objective response rate [ORR] in patients with measurable disease) were examined for patient subgroups corresponding to each of the identified significant prognostic factors. Subpopulation treatment effect pattern plot (STEPP) methodology was performed to examine the association between treatment-free interval (TFI) following adjuvant endocrine therapy and outcomes of endocrine therapy alone or in combination with abemaciclib in MONARCH 3. Results: Analyses of clinical factors in over 1000 patients confirmed the following to have prognostic value: bone-only disease, liver metastases, tumor grade, progesterone receptor (PgR) status, and ECOG performance status. Prognosis was poor in patients with liver metastases, PgR-negative tumors, and high-grade tumors. While all subpopulations benefited from the addition of abemaciclib to endocrine therapy regardless of prognosis, substantial benefit of abemaciclib was observed in poor prognosis subgroups, characterized by large increases in PFS (hazard ratios = 0.4 to 0.5) and ORR (over 30%). In addition, STEPP analysis of TFI on a subset of the MONARCH 3 population showed that patients with the shortest TFI appeared to have a poorer prognosis and received more benefit from the addition of abemaciclib compared to patients with longer TFI. Conclusions: This exploratory analysis has provided data that could help optimize treatment strategies by identifying that patients with poor prognostic factors may receive greater benefit from the addition of abemaciclib to endocrine therapy. Citation Format: Goetz MP, O'Shaughnessy J, Sledge Jr. GW, Martin M, Lin Y, Forrester T, Mockbee C, Smith IC, Di Leo A, Johnston S. The benefit of abemaciclib in prognostic subgroups: An exploratory analysis of combined data from the MONARCH 2 and 3 studies [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr GS6-02.

Age does not influence efficacy of ramucirumab in advanced gastric cancer: Subgroup analyses of REGARD and RAINBOW.

REGARD and RAINBOW were global, phase 3, randomized, double-blind trials of second-line ramucirumab for metastatic gastric or gastroesophageal junction adenocarcinoma. Exploratory subgroup analyses were described to assess the efficacy and safety of ramucirumab in REGARD and RAINBOW in young (≤45 and <65years) and elderly (≥65, ≥70, and ≥75years) patients. Patients were randomized 2:1 to receive ramucirumab plus best supportive care or placebo plus best supportive care (REGARD) or 1:1 to ramucirumab plus paclitaxel or placebo plus paclitaxel (RAINBOW). Subpopulation Treatment Effect Pattern Plots assessed efficacy and adverse events by age groups for ramucirumab versus placebo. The hazard ratios (HRs) for overall survival favored treatment with ramucirumab: REGARD ≤45years (HR: 0.59, 95% confidence interval: 0.27-1.26), <65years (0.80, 0.59-1.10), ≥65years (0.72, 0.48-1.08), ≥70years (0.73, 0.44-1.23), and ≥75years (0.59, 0.25-1.37); and RAINBOW ≤45 years (0.56, 0.33-0.93), <65years (0.78, 0.63-0.97), ≥65years (0.88, 0.66-1.18), and ≥70years (0.88, 0.60-1.28). The exception was elderly patients aged ≥75years in RAINBOW (0.97, 0.47-2.01); however, patient numbers were low in this subgroup (n=36). Similar findings were observed for progression-free survival, for which HRs numerically favored ramucirumab-treated patients. Adverse events (including grade ≥3) were not associated with age. In comparison with placebo, ramucirumab conferred improvements in efficacy across age groups with a tolerable safety profile. Despite some limitations, these exploratory analyses support the use of ramucirumab in advanced gastric cancer, irrespective of age.

171P - Composite index of risk shows that benefit from adjuvant dose dense chemotherapy is not confined to triple negative breast cancer

Background: Compared with the standard interval adjuvant chemotherapy, dose-dense schedule is proved to increase disease-free survival (DFS) in node-positive early breast cancer (EBC) patients (pts). To date, GIM2 is the only trial supporting the role of dose-dense chemotherapy in pts with hormone receptor-negative (HR-) or hormone receptor-positive tumours (HR+) (Del Mastro et al. Lancet 2015). To further refine the evidence of treatment effect in the HR+ subgroup, a composite index of risk was developed including clinico-pathological features. Methods: The randomized phase III GIM2 trial enrolled 2091 pts with node-positive EBC (primary endpoint: DFS). A continuous, composite measure of treatment benefit was determined from a Cox model incorporating potential predictive factors (age: 25-40/41-55/56-71; histological grade: 1 + 2/3; HR status: positive/negative; ki-67 levels: ≤20%/>20%). Subpopulation treatment effect pattern plot methodology was used to reveal differential treatment effects on DFS according to composite index. The study analyzed the cohort of pts with HER2- (N = 1127) disease with a special focus on HR+ disease (N = 980). Results: On average, the magnitude of benefit with dose dense chemotherapy versus standard chemotherapy widely varied according to composite measure of specific features. In the HER2- subgroup, the highest benefit was observed in pts with G3, HR-, >10 positive nodes, age <40 yrs, ki-67 > 20% (hazard ratio for DFS 0.57, 95% CI 0.35-0.94). Notably, among pts with HR+ disease, the following clinic-pathological characteristics conferred the highest benefit: G3, ≥4 positive nodes, age ≥56 yrs, ki-67 > 20% (hazard ratio for DFS 0.66, 95% CI 0.38-1.15). Conclusions: Composite risk evaluation and corresponding subpopulation treatment effect pattern plot methodology suggest that benefit of dose dense adjuvant chemotherapy is not confined to triple negative EBC. Legal entity responsible for the study: Gruppo Italiano Mammella Funding: None Disclosure: All authors have declared no conflicts of interest.

Disseminated tumour cells with highly aberrant genomes are linked to poor prognosis in operable oesophageal adenocarcinoma

Background:Chromosomal instability (CIN) has repeatedly been identified as a prognostic marker. Here we evaluated the percentage of aberrant genome per cell (PAG) as a measure of CIN in single disseminated tumour cells (DTC) isolated from patients with operable oesophageal adenocarcinoma (EAC), to assess the impact of CINhigh DTCs on prognosis.Methods:We isolated CK18positive DTCs from bone marrow (BM) or lymph node (LN) preparations of operable EAC patients. After whole-genome amplification, single DTCs were analysed for chromosomal gains and losses using metaphase-based comparative genomic hybridisation (mCGH). We calculated the PAG for each DTC and determined the critical threshold value that identifies high-risk patients by STEPP (Subpopulation Treatment Effect Pattern Plot) analysis in two independent EAC patient cohorts (cohort #1, n=44; cohort #2; n=29).Results:The most common chromosomal alterations observed among the DTCs were typical for EAC, but the DTCs showed a varying PAG between individual patients. Generally, LNDTCs displayed a significantly higher PAG than BMDTCs. STEPP analysis revealed an increasing PAG of DTCs to be correlated with an increased risk for short survival in two independent EAC cohorts as well as in the corresponding pooled analysis. In all three data sets (cohort #1, cohort #2 and pooled cohort), PAGhigh DTCs conferred an independent risk for a significantly decreased survival.Conclusions:The analysis of PAG/CIN in solitary marker-positive DTCs identifies operable EAC patients with poor prognosis, indicating a more aggressive minimal residual disease.

Prognostic impact of proliferation for resected early stage ‘pure’ invasive lobular breast cancer: Cut-off analysis of Ki67 according to histology and clinical validation

IntroductionThe intent of this analysis was to investigate and validate the prognostic potential of Ki67 in a multi-center series of patients affected by early stage ‘pure’ invasive lobular carcinoma (ILC). MethodsClinical-pathological data of patients affected by ILC were correlated with overall survival and disease-free survival (OS/DFS); data from a parallel invasive ductal carcinoma (IDC) patients' cohort were gathered as well. The maximally selected Log-Rank statistics analysis was applied to Ki67 continuous variable to estimate the appropriate cut-off. The Subpopulation Treatment Effect Pattern Plot (STEPP) analysis was performed as well. ResultsData from overall 1097 (457/222 ILC: training/validation set; 418 IDC) patients were gathered. The identified optimal Ki67 cut-offs were 4% and 14% for DFS in ILC and IDC cohort, respectively. In ILC patients, the Ki67 cut-off was an independent OS predictor. Ten-years OS and DFS were 89.9% and 77.2% (p = 0.007) and 79.4% and 69.2% (p = 0.03) for patients with Ki67 ≤ 4% and >4%, respectively. In IDC patients, 10-years OS was 93.8% and 71.7%, p = 0.02, DFS was 84.0% and 52.6%, p = 0.0003, for patients with Ki67 ≤ 14% and >14%, respectively. In the validation set, the optimal Ki67 OS cut-off was 5%. The STEPP analysis showed that in the presence of low Ki67 values, IDC patients have a better DFS than ILC patients, while with the increase of values the prognosis tends to overlap. ConclusionsDespite the retrospective design of the study, the prognostic relevance of Ki67 (as well as its optimal cut-off) seems to significantly differ according to breast cancer histology.

Abstract P1-09-12: Dose dense adjuvant chemotherapy in patients with early breast cancer: Differential treatment effects according to composite index of benefit

Abstract Background In patients with node-positive early breast cancer (EBC), dose-dense adjuvant chemotherapy improves disease-free survival (DFS) compared with standard interval chemotherapy. The GIM2 trial supports the value of dose-dense chemotherapy and suggests that the benefit is present in patients with hormone receptor-negative or hormone receptor-positive tumours (Del Mastro et al. Lancet 2015). In order to individualize decision making, there is a need to examine the absolute treatment effects of dose dense chemotherapy according to patient and tumor characteristics. Patients and Methods The randomized phase III GIM2 trial enrolled 2091 patients. The primary endpoint was DFS. A continuous, composite measure of treatment benefit for each patient was determined from a Cox model incorporating potential predictive factors (age: 25-40/41-55/56-71; histological grade: 1+2/3; hormonal receptor status: positive/negative). Subpopulation treatment effect pattern plot methodology was used to reveal differential treatment effects on DFS according to composite index. The study focused on patients with HER2–negative disease (N=1287). Results On average, the magnitude of benefit with dose dense chemotherapy versus standard chemotherapy ranged widely across different subpopulations, as quantified by the composite measure of relevant variables. The highest benefit was observed in patients with high grade, hormone receptor-negative disease (hazard ratio for DFS 0.44, 95% CI 0.23-0.83). Of note, a relevant benefit was observed also in patients with high grade, hormone receptor-positive disease (hazard ratio for DFS 0.74, 95% CI 0.50-1.09). Conclusion The absolute improvement in DFS with dose dense adjuvant chemotherapy is substantial in some patients with node-positive HER2-negative breast cancer, particularly those regarded as having high index risk (hormone receptor-negative, high grade disease). Interestingly, a significant effect of dose dense chemotherapy was observed also in patients with hormone receptor-positive, high grade disease. Citation Format: Puglisi F, Ceppi M, Cognetti F, De Placido S, Bruzzi P, De Laurentiis M, Bisagni G, Cavazzini G, Durando A, Turletti A, Valle E, Montemurro F, Barni S, Ardizzoni A, Gamucci T, Colantuoni G, Del Mastro L. Dose dense adjuvant chemotherapy in patients with early breast cancer: Differential treatment effects according to composite index of benefit [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P1-09-12.

Efficacy and safety of ramucirumab (RAM) for metastatic gastric or gastroesophageal junction (GEJ) adenocarcinoma across age subgroups in two global phase 3 trials.

3 Background: REGARD and RAINBOW are two phase 3 studies which demonstrated significant survival benefits and manageable toxicity in patients with advanced gastric cancer (GC) randomized in the second-line setting to receive RAM or placebo. We examined outcomes by age (≤45, 45-70, ≥70, ≥75 years). Methods: Patients were randomized 2:1 to receive RAM 8 mg/kg + best-supportive care (BSC) or placebo (PL) + BSC (REGARD); or 1:1 RAM 8 mg/kg + paclitaxel (PTX) or PL + PTX (RAINBOW). Kaplan-Meier analysis and Cox proportional hazards regression were performed for overall survival (OS) and progression-free survival (PFS). Subpopulation treatment effect pattern plot (STEPP) assessed efficacy and the incidence of adverse events (AEs) across age subgroups. Results: Baseline characteristics were generally well balanced between arms amongst the age subgroups. Efficacy outcomes are summarized in the Table. STEPP analysis revealed no obvious patterns for differential risks in terms of efficacy and AEs (any grade or grade ≥3) according to age. Conclusions: Compared with PL, the efficacy of RAM was maintained in all age groups, with similar rates of toxicity. Despite some limitations regarding patient numbers in some age subgroups, this exploratory subgroup analysis supports the use of RAM for the treatment of GC, irrespective of age. Clinical trial information: NCT00917384 and NCT01170663. [Table: see text]

Risk-based evaluation of efficacy of rolofylline in patients hospitalized with acute heart failure — Post-hoc analysis of the PROTECT trial

BackgroundThe selective adenosine A1 receptor antagonist rolofylline showed a neutral overall result on clinical outcomes in the PROTECT trial. However, we hypothesized that response to rolofylline treatment could be influenced by underlying clinical risk. MethodsWe performed a post-hoc analysis of the PROTECT trial — a large, double-blind, randomized, placebo-controlled trial that enrolled 2033 patients. Baseline risk of 180-day all-cause mortality was estimated using a previously published 8-item model. Evaluation of efficacy of rolofylline across subpopulations defined based on estimated risk of mortality was performed using subpopulation treatment effect pattern plot (STEPP) analysis. Findings were validated in an independent cohort of acute heart failure patients. ResultsMedian estimated risk of mortality was 13.0%, IQR [8.0%–23.0%] and was comparable between the rolofylline and placebo arms. In low to intermediate risk subgroups of patients, rolofylline was associated with a higher rate of 180-day all-cause mortality (11.9% in the rolofylline versus 8.4% in the placebo arms, p=0.050). In the high risk subgroup of patients, particularly those with estimated risk of mortality between 20% and 30%, 180-day all-cause mortality rate was markedly lower in the rolofylline arm (18.4% in the rolofylline versus 34.0% in the placebo arms, p=0.003). The trend towards potential harm with rolofylline treatment in the low to intermediate risk subpopulations and significant benefit in high risk patients was also observed in the validation cohort. ConclusionOur findings suggest that selective adenosine A1 receptor antagonism could be harmful in low risk acute heart failure patients, while it might significantly benefit higher risk patients.

Absolute Benefit of Adjuvant Endocrine Therapies for Premenopausal Women With Hormone Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Negative Early Breast Cancer: TEXT and SOFT Trials.

Risk of recurrence is the primary consideration in breast cancer adjuvant therapy recommendations. The TEXT (Tamoxifen and Exemestane Trial) and SOFT (Suppression of Ovarian Function Trial) trials investigated adjuvant endocrine therapies for premenopausal women with hormone receptor-positive breast cancer, testing exemestane plus ovarian function suppression (OFS), tamoxifen plus OFS, and tamoxifen alone. We examined absolute treatment effect across a continuum of recurrence risk to individualize endocrine therapy decision making for premenopausal women with human epidermal growth factor receptor 2 (HER2) -negative disease. The TEXT and SOFT hormone receptor-positive, HER2-negative analysis population included 4,891 women. The end point was breast cancer-free interval (BCFI), defined as time from random assignment to first occurrence of invasive locoregional, distant, or contralateral breast cancer. A continuous, composite measure of recurrence risk for each patient was determined from a Cox model incorporating age, nodal status, tumor size and grade, and estrogen receptor, progesterone receptor, and Ki-67 expression levels. Subpopulation treatment effect pattern plot methodology revealed differential treatment effects on 5-year BCFI according to composite risk. SOFT patients who remained premenopausal after chemotherapy experienced absolute improvement of 5% or more in 5-year BCFI with exemestane plus OFS versus tamoxifen plus OFS or tamoxifen alone, reaching 10% to 15% at intermediate to high composite risk; the benefit of tamoxifen plus OFS versus tamoxifen alone was apparent at the highest composite risk. The SOFT no-chemotherapy cohort-for whom composite risk was lowest on average-did well with all endocrine therapies. For TEXT patients, the benefit of exemestane plus OFS versus tamoxifen plus OFS in 5-year BCFI ranged from 5% to 15%; patients not receiving chemotherapy and with lowest composite risk did well with both treatments. Premenopausal women with hormone receptor-positive, HER2-negative disease and high recurrence risk, as defined by clinicopathologic characteristics, may experience improvement of 10% to 15% in 5-year BCFI with exemestane plus OFS versus tamoxifen alone. An improvement of at least 5% may be achieved for women at intermediate risk, and improvement is minimal for those at lowest risk.

Evaluating Markers for Guiding Treatment.

The subpopulation treatment effect pattern plot (STEPP) is an appealing method for assessing the clinical impact of a predictive marker on patient outcomes and identifying a promising subgroup for further study. However, its original formulation lacked a decision analytic justification and applied only to a single marker. We derive a decision-analytic result that motivates STEPP. We discuss the incorporation of multiple predictive markers into STEPP using risk difference, cadit, and responders-only benefit functions. Applying STEPP to data from a breast cancer treatment trial with multiple markers, we found that none of the three benefit functions identified a promising subgroup for further study. Applying STEPP to hypothetical data from a trial with 100 markers, we found that all three benefit functions identified promising subgroups as evidenced by the large statistically significant treatment effect in these subgroups. Because the method has desirable decision-analytic properties and yields an informative plot, it is worth applying to randomized trials on the chance there is a large treatment effect in a subgroup determined by the predictive markers.

Prognostic and predictive value of Ki-67 in triple-negative breast cancer.

This study was to investigate the prognostic role of Ki-67 in further classification of triple negative breast cancer (TNBC), and to test whether high expression level of Ki67 can predict benefit from carboplatin. From January 2004 to December 2012, 363 patients operated for TNBC were identified through the institutional clinical database. After a median follow-up time of 34 months (5.2–120.0 months), 62 patients (17.1%) had relapses and 33 patients (9.1%) died of breast cancer. In univariate analysis, high Ki-67 index as well as larger tumor size and lymph node involvement was associated with shorter disease-free survival (DFS) and overall survival (OS). In multivariate analysis, high Ki-67 is an independent risk factor for DFS (Risk Ratio, RR: 2.835, 95% confidence interval, 95% CI: 1.586–5.068, P < 0.001) and OS (RR: 3.180, 95% CI: 1.488–6.793, P = 0.003). When analyzing the 3-year DFS by Ki-67 distribution, Subpopulation Treatment Effect Pattern Plot analysis showed a beneficial effect of carboplatin in patients with high Ki-67 index. In conclusion, TNBC is probably a heterogeneous disease with different characteristics and prognosis, and may be further subdivided according to the Ki-67 expression levels. Patients in the high Ki- 67 group seem to benefit more from treatment with carboplatin, but this needs to be further verified.

Subpopulation Treatment Effect Pattern Plot (STEPP) analysis for continuous, binary, and count outcomes.

For the past few decades, randomized clinical trials have provided evidence for effective treatments by comparing several competing therapies. Their successes have led to numerous new therapies to combat many diseases. However, since their conclusions are based on the entire cohort in the trial, the treatment recommendation is for everyone, and may not be the best option for an individual. Medical research is now focusing more on providing personalized care for patients, which requires investigating how patient characteristics, including novel biomarkers, modify the effect of current treatment modalities. This is known as heterogeneity of treatment effects. A better understanding of the interaction between treatment and patient-specific prognostic factors will enable practitioners to expand the availability of tailored therapies, with the ultimate goal of improving patient outcomes. The Subpopulation Treatment Effect Pattern Plot (STEPP) approach was developed to allow researchers to investigate the heterogeneity of treatment effects on survival outcomes across values of a (continuously measured) covariate, such as a biomarker measurement. Here, we extend the Subpopulation Treatment Effect Pattern Plot approach to continuous, binary, and count outcomes, which can be easily modeled using generalized linear models. With this extension of Subpopulation Treatment Effect Pattern Plot, these additional types of treatment effects within subpopulations defined with respect to a covariate of interest can be estimated, and the statistical significance of any observed heterogeneity of treatment effect can be assessed using permutation tests. The desirable feature that commonly used models are applied to well-defined patient subgroups to estimate treatment effects is retained in this extension. We describe a simulation study to confirm that the proper Type I error rate is maintained when there is no treatment heterogeneity, and a power study to show that the statistics have power to detect treatment heterogeneity under alternative scenarios. As an illustration, we apply the methods to data from the Aspirin/Folate Polyp Prevention Study, a clinical trial evaluating the effect of oral aspirin, folic acid, or both as a chemoprevention agent against colorectal adenomas. The pre-existing R software package stepp has been extended to handle continuous, binary, and count data using Gaussian, Bernoulli, and Poisson models, and it is available on the Comprehensive R Archive Network. The extension of the method and the availability of new software now permit STEPP to be applied to the full range of clinical trial end points.

Meta-STEPP: subpopulation treatment effect pattern plot for individual patient data meta-analysis.

We have developed a method, called Meta-STEPP (subpopulation treatment effect pattern plot for meta-analysis), to explore treatment effect heterogeneity across covariate values in the meta-analysis setting for time-to-event data when the covariate of interest is continuous. Meta-STEPP forms overlapping subpopulations from individual patient data containing similar numbers of events with increasing covariate values, estimates subpopulation treatment effects using standard fixed-effects meta-analysis methodology, displays the estimated subpopulation treatment effect as a function of the covariate values, and provides a statistical test to detect possibly complex treatment-covariate interactions. Simulation studies show that this test has adequate type-I error rate recovery as well as power when reasonable window sizes are chosen. When applied to eight breast cancer trials, Meta-STEPP suggests that chemotherapy is less effective for tumors with high estrogen receptor expression compared with those with low expression. Copyright © 2016 John Wiley & Sons, Ltd.

Subpopulation Treatment Effect Pattern Plot (STEPP) analysis of Ki67 assay according to histology: prognostic relevance for resected early stage 'pure' and 'mixed' lobular breast cancer.

BackgroundThe aim of this analysis was to investigate the potential impact of Ki67 assay in a series of patients affected by early stage invasive lobular carcinoma (ILC) undergone surgery.MethodsClinical-pathological data were correlated with disease-free and overall survival (DFS/OS). The maximally selected Log-Rank statistics analysis was applied to the Ki67 continuous variable to estimate appropriate cut-offs. The Subpopulation Treatment Effect Pattern Plot (STEPP) analysis was performed to assess the interaction between ‘pure’ or ‘mixed’ histology ILC and Ki67.ResultsAt a median follow-up of 67 months, 10-years DFS and OS of 405 patients were 67.8 and 79.8 %, respectively. Standardized Log-Rank statistics identified 2 optimal cut-offs (6 and 21 %); 10-years DFS and OS were 75.1, 66.5, and 30.2 % (p = 0.01) and 84.3, 76.4 and 59 % (p = 0.003), for patients with a Ki67 < 6 %, between 6 and 21 %, and >21 %, respectively. Ki67 and lymph-node status were independent predictor for longer DFS and OS at the multivariate analysis, with radiotherapy (for DFS) and age (for OS). Ki67 highly replicated at the internal cross-validation analysis (DFS 85 %, OS 100 %). The STEPP analysis showed that DFS rate decreases as Ki67 increases and those patients with ‘pure’ ILC performed worse than ‘mixed’ histology.ConclusionsDespite the retrospective and exploratory nature of the study, Ki67 was able to significantly discriminate the prognosis of patients with ILC, and the effect was more pronounced for patients with ‘pure’ ILC.Electronic supplementary materialThe online version of this article (doi:10.1186/s13046-016-0325-z) contains supplementary material, which is available to authorized users.

Identifying treatment effect heterogeneity in clinical trials using subpopulations of events: STEPP.

Investigators conducting randomized clinical trials often explore treatment effect heterogeneity to assess whether treatment efficacy varies according to patient characteristics. Identifying heterogeneity is central to making informed personalized healthcare decisions. Treatment effect heterogeneity can be investigated using subpopulation treatment effect pattern plot (STEPP), a non-parametric graphical approach that constructs overlapping patient subpopulations with varying values of a characteristic. Procedures for statistical testing using subpopulation treatment effect pattern plot when the endpoint of interest is survival remain an area of active investigation. A STEPP analysis was used to explore patterns of absolute and relative treatment effects for varying levels of a breast cancer biomarker, Ki-67, in the phase III Breast International Group 1-98 randomized clinical trial, comparing letrozole to tamoxifen as adjuvant therapy for postmenopausal women with hormone receptor-positive breast cancer. Absolute treatment effects were measured by differences in 4-year cumulative incidence of breast cancer recurrence, while relative effects were measured by the subdistribution hazard ratio in the presence of competing risks using O-E (observed-minus-expected) methodology, an intuitive non-parametric method. While estimation of hazard ratio values based on O-E methodology has been shown, a similar development for the subdistribution hazard ratio has not. Furthermore, we observed that the subpopulation treatment effect pattern plot analysis may not produce results, even with 100 patients within each subpopulation. After further investigation through simulation studies, we observed inflation of the type I error rate of the traditional test statistic and sometimes singular variance-covariance matrix estimates that may lead to results not being produced. This is due to the lack of sufficient number of events within the subpopulations, which we refer to as instability of the subpopulation treatment effect pattern plot analysis. We introduce methodology designed to improve stability of the subpopulation treatment effect pattern plot analysis and generalize O-E methodology to the competing risks setting. Simulation studies were designed to assess the type I error rate of the tests for a variety of treatment effect measures, including subdistribution hazard ratio based on O-E estimation. This subpopulation treatment effect pattern plot methodology and standard regression modeling were used to evaluate heterogeneity of Ki-67 in the Breast International Group 1-98 randomized clinical trial. We introduce methodology that generalizes O-E methodology to the competing risks setting and that improves stability of the STEPP analysis by pre-specifying the number of events across subpopulations while controlling the type I error rate. The subpopulation treatment effect pattern plot analysis of the Breast International Group 1-98 randomized clinical trial showed that patients with high Ki-67 percentages may benefit most from letrozole, while heterogeneity was not detected using standard regression modeling. The STEPP methodology can be used to study complex patterns of treatment effect heterogeneity, as illustrated in the Breast International Group 1-98 randomized clinical trial. For the subpopulation treatment effect pattern plot analysis, we recommend a minimum of 20 events within each subpopulation.

Impact of body mass index on neoadjuvant treatment outcome: a pooled analysis of eight prospective neoadjuvant breast cancer trials.

Obesity is associated with an increased risk of breast cancer (BC) and poorer outcome. We assessed the impact of body mass index (BMI) on pathological complete response (pCR), disease-free (DFS), and overall survival (OS), according to BC subtypes in patients with primary BC treated with neoadjuvant chemotherapy. 8,872 patients with primary BC from eight neoadjuvant trials were categorized according to BMI: underweight (<18.5kg/m(2)), normal weight (18.5 to <25kg/m(2)), overweight (25 to <30kg/m(2)), obese (30 to <40kg/m(2)), and very obese (≥40kg/m(2)). BC subtypes were defined as luminal-like (ER/PgR-positive and HER2-negative), HER2/luminal (ER/PgR-positive and HER2-positive), HER2-like (ER/PgR-negative and HER2-positive), and triple-negative (TNBC; ER/PgR- and HER2-negative). pCR rate was higher in normal weight patients compared with all other BMI groups (P=0.003). Mean DFS and OS were shorter in obese (87.3months, P=0.014 and 94.9months, P=0.001, respectively) and very obese (66.6months, P<0.001 and 75.3months, P<0.001, respectively) compared with normal weight patients (91.5 and 98.8months, respectively) which was confirmed by subpopulation treatment effect pattern plot analyses and was consistent in luminal-like and TNBC. No interaction was observed between BMI and pCR. Normal weight patients experienced less non-hematological adverse events (P=0.002) and were more likely to receive full taxane doses (P<0.001) compared with all other BMI groups. In multivariable analysis, the dose of taxanes was predictive for pCR (P<0.001). Higher BMI was associated with lower pCR and a detrimental impact on survival. Normal weight patients had the best compliance to chemotherapy and received the highest taxane doses, which seems to be related with treatment outcomes.

Lorenz Curves and Treatment-Covariate Interactions in Clinical Trials

A common objective in comparative two-treatment randomized clinical trials is the study of the possible heterogeneity of the treatment effect across subgroups of patients, with the objective of identifying patients who benefit the most (or the least) from a new treatment. Here we describe the connection that exists between an exploratory approach to such problem (STEPP, or the Subpopulation Treatment Effect Pattern Plot approach) and the Lorenz curve, and in particular the generalized Lorenz curve. We exploit such connection to construct a test for the absence of interaction between a continuous covariate and the difference in the mean of a continuous outcome between the two treatment groups. We also review some recent developments in the study of concentration for right censored survival data, which are also closed related to the Lorenz curve. DOI: http://dx.doi.org/10.4038/sljastats.v5i4.7788