Clinico-Pathological Features Influencing the Prognostic Role of Body Mass Index in Patients With Advanced Renal Cell Carcinoma Treated by Immuno-Oncology Combinations (ARON-1)

BACKGROUND

Obesity has been associated with improved response to immunotherapy in cancer patients. We investigated the role of body mass index (BMI) in patients from the ARON-1 study (NCT05287464) treated by dual immuno-oncology agents (IO+IO) or a combination of immuno-oncology drug and a tyrosine kinase inhibitors (TKI) as first-line therapy for metastatic renal cell carcinoma (mRCC).

PATIENTS AND METHODS

Medical records of patients with documented mRCC treated by immuno-oncology combinations were reviewed at 47 institutions from 16 countries. Patients were assessed for overall survival (OS), progression-free survival (OS), and overall clinical benefit (OCB), defined as the sum of the rate of partial/complete responses and stable disease. Univariate and multivariate analyses were used to explore the association of variables of interest with survival.

RESULTS

A total of 675 patients were included; BMI was >25 kg/m2 in 345 patients (51%) and was associated with improved OS (55.7 vs. 28.4 months, P < .001). The OCB of patients with BMI >25 kg/m2 versus those with BMI ≤25 kg/m2 was significantly higher only in patients with nonclear cell histology (81% vs. 65%, P = .011), and patients with liver metastases (76% vs. 58%, P = .007), Neutrophil to lymphocyte ratio >4 (77% vs 62%, P = .022) or treated by nivolumab plus ipilimumab (77% vs. 64%, P = .044). In the BMI ≤25 kg/m2 subgroup, significant differences were found between patients with NLR >4 versus ≤4 (62% vs. 82%, P = .002) and patients treated by IO+IO versus IO+TKIs combinations (64% vs. 83%, P = .002).

CONCLUSION

Our study suggests that the prognostic significance and the association of BMI with treatment outcome varies across clinico-pathological mRCC subgroups.

Additional biological and clinical characteristics to refine International Metastatic RCC Database Consortium (IMDC) prognostic/predictive assessment in metastatic renal cell carcinoma (mRCC)

705

Background: mRCC prognostic stratification is currently based on the IMDC score. However, some patients experience outcomes markedly different from those predicted by the median outcome of their risk class. Moreover, risk scores are prognostic, rather than predictive, and developing reproducible, affordable, biology-based predictive biomarkers to tailor treatment choices in individual pts remains an unmet clinical need. Methods: A retrospective cohort of 113 mRCC pts treated at the Verona University Hospital Trust between 2013 and 2021 was explored to identify additional clinical prognostic factors for OS to complement IMDC score. Outlier pts (i.e. individual pts whose clinical outcome differed significantly from the median of the IMDC group they belonged to) were further explored to find putative molecular signatures, using FISH and IHC assays. Results: At a median follow-up of 69 months (range 18-136m), novel variables impacting on OS in addition to IMDC risk group were identified: bone, central nervous system (CNS) and pancreatic mets and high neutrophil/lymphocyte ratio (N/L) with a cut-off of 3.2. We also analyzed outliers: 3 good-risk pts with lower than expected OS, 4 poor-risk pts with longer than expected OS, and two intermediate-risk pts with long treatment response. Two putative signatures were found predictive of disease behavior: a cold signature (9p loss, poor of TILs) and a hot signature (rich in CD56+, CD15+ and CD8+ infiltrating cells). The cold and hot signatures were found in good risk pts with bad OS and poor-risk pts good OS, respectively; both intermediate-risk pts had a hot signature, consistent with their disease course. Conclusions: Additional clinical and molecular factors have been identified to possibly improve IMDC prognostic performance; multivariable models including them along with standard IMDC are being developed. Molecular analysis suggests potential signatures to be applied in routine clinical practice, whose predictive performance could be validated in prospective multicentric trials. [Table: see text]

Interleukin-8 in Colorectal Cancer: A Systematic Review and Meta-Analysis of Its Potential Role as a Prognostic Biomarker

Among soluble actors that have emerged as druggable factors, the chemokine interleukin-8 (IL-8) has emerged as a possible determinant of response to immunotherapy and targeted treatment in several cancer types; however, its prognostic/predictive role in colorectal cancer (CRC) remains to be established. We: (i) conducted a systematic review of published literature on IL-8 expression in CRC; (ii) searched public transcriptomics databases; (iii) investigated IL-8 expression, by tumor and infiltrating cells, in a series of CRC samples; and (iv) carried out a meta-analysis of published literature correlating IL-8 expression and CRC prognosis. IL-8 possesses an important role as a mediator of the bidirectional crosstalk between tumor/stromal cells. Transcriptomic analysis indicated that specific IL-8 transcripts were significantly overexpressed in CRC compared to normal colon mucosa. Moreover, in our series we observed a statistically significant correlation between PTEN-loss and IL-8 expression by infiltrating mononuclear and tumor cells. In total, 12 papers met our meta-analysis inclusion criteria, demonstrating that high IL-8 levels significantly correlated with shorter overall survival and progression-free survival. Sensitivity analysis demonstrated a highly significant correlation with outcome for circulating, but not for tissue-detected, IL-8. IL-8 is overexpressed in CRC tissues and differentially produced by tumor or stromal components depending on CRC genetic background. Moreover, circulating IL-8 represents a strong prognostic factor in CRC, suggesting its use in the refining of prognostic CRC assessment and potentially the tailoring of therapeutic strategies in individual CRC patients.

High Levels of Circulating Monocytic Myeloid-Derived Suppressive-Like Cells Are Associated With the Primary Resistance to Immune Checkpoint Inhibitors in Advanced Non-Small Cell Lung Cancer: An Exploratory Analysis

Background

Immunotherapy has become the standard of care for non-small cell lung cancer (NSCLC) patients. Some patients experience primary resistance to immunotherapy. Currently, we lack a marker of resistance to immunotherapy. Myeloid-derived suppressive-like cells (MDSCs) can reduce tumor response rate and survival outcomes.

Methods

This is an exploratory prospective observational study on metastatic NSCLC patients starting immunotherapy. Baseline peripheral blood samples were collected. Monocytic (M)-MDSCs were analyzed by flow cytometry. The main clinical outcomes were tumor response, progression-free survival (PFS), and overall survival (OS). The association between MDSC levels and tumor response was assessed. The association of PFS with OS was investigated using the Kaplan–Meier method and the Cox proportional hazards model.

Results

Twenty-two patients were included. The median M-MDSC value was higher in patients with progressive disease than patients with stable disease or partial response, p = 0.045. The median MDSC value in the overall population was 1.9. We found worse PFS (HR = 2.51; p = 0.046) and OS (HR = 2.68; p = 0.042) in patients with M-MDSC values higher than the median.

Conclusions

In this exploratory analysis, high M-MDSC levels are strongly associated with primary resistance to immunotherapy. If validated in larger studies, MDSC levels in blood samples could help to select NSCLC patients for higher benefit from immunotherapy.

Early primary tumor response in metastatic RCC patients treated with immune checkpoint inhibitors-based combinations

349

Background: 25-30% of renal cell carcinoma presents with metastases (mRCC) at diagnosis. The activity of immune checkpoint inhibitor (ICI)-combinations on the primary tumor (PT) is debated. Patients and Methods: mRCC patients (pts) with PT who received first-line nivolumab plus ipilimumab (N/I) or pembrolizumab plus axitinib (P/A) were included. We investigated the early primary tumor response (EPTR) at the first radiological assessment. Results: 73 pts were included. The median early reduction of the PT longest diameter was 12.4% with P/A versus 6.2% with N/I (p = 0.42). We evaluated if the type of EPTR could affect the metastases response. Among pts with PT stable disease (SD), 8.3% had metastatic disease progression (PD) with P/A and 34.8% with N/I. Early PT partial response (PR) was associated with no metastatic PD with both N/I and P/A. The 2 pts with PT PD had also metastatic PD to P/A. Of the 3 PT with PD to N/I, 1 had metastatic SD and 2 PD. In the overall population, of the 94.1% without PT progression (PR+SD), 47.5% had metastatic PR, 35.6% SD, 16.9% PD. Conclusions: ICIs-combinations achieved an early PT PR in about 10-20%, without any complete responses. Only a small percentage of PT had an early PD, mainly associated with metastatic PD. However, among those PT without an early progression, metastatic PR can be achieved in approximately 50% of cases.[Table: see text]

Nivolumab in Combination with Stereotactic Body Radiotherapy in Pretreated Patients with Metastatic Renal Cell Carcinoma. Results of the Phase II NIVES Study

BACKGROUND

Nivolumab showed an overall survival (OS) benefit in pretreated metastatic renal cell carcinoma (mRCC). The role of stereotactic body radiotherapy (SBRT) in mRCC remains to be defined.

OBJECTIVE

Our aim was to evaluate the efficacy and safety of SBRT in combination with nivolumab in second- and third-line mRCC patients.

DESIGN, SETTING, AND PARTICIPANTS

The NIVES study was a phase II, single-arm, multicenter trial in patients with mRCC with measurable metastatic sites who progressed after antiangiogenic therapy, of whom at least one was suitable for SBRT.

INTERVENTION

The patients received SBRT to a lesion at a dose of 10 Gy in three fractions for 7 d from the first infusion of nivolumab. Nivolumab was given at an initial dose of 240 mg every 14 d for 6 mo and then 480 mg q4-weekly in responding patients.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

We hypothesized that nivolumab plus SBRT improves the objective response rate (ORR) compared with nivolumab alone from 25% (derived from historical controls) to 40%. Secondary endpoints were progression-free survival (PFS), OS, disease control rate (DCR) of irradiated and nonirradiated metastases, and safety.

RESULTS AND LIMITATIONS

Sixty-nine patients were enrolled from July 2017 to March 2019. The ORR was 17% and the DCR was 55%. The median PFS was 5.6 mo (95% confidence interval [CI], 2.9-7.1) and median OS 20 mo (95% CI, 17-not reached). After 1.5 yr of follow-up, 23 patients died. The median time to treatment response was 2.8 mo and median duration of response was 14 mo. No new safety concerns arose.

CONCLUSIONS

We did not find sufficient evidence to suggest that nivolumab in combination with SBRT provides an added benefit in pretreated mRCC patients; it should however be evaluated in patients with oligometastatic or oligoprogressive disease.

PATIENT SUMMARY

Nivolumab in combination with stereotactic body radiotherapy does not provide evidence of increased outcomes in metastatic renal cell carcinoma patients. However this approach was safe and showed a good response of the irradiated lesions.

Programmed death ligand-1 (PD-L1) expression in patients (pts) with metastatic renal cell carcinoma (mRCC) treated with nivolumab (NIVO) in combination with stereotactic body radiotherapy (SBRT) in NIVES study

4558

Background: The NIVES study represents the first prospective trial with NIVO in combination with SBRT in pre-treated mRCC patients. This study did not meet the primary endpoint in terms of objective response rate (ORR) as previously reported. However this combination showed a faster time to treatment response, a long progression free survival and median duration of response without increasing toxicities. Here we have tested with an exploratory analysis the correlation between PD-L1 expression and clinical outcomes in pts treated with NIVO plus SBRT. Methods: PD-L1 expression was assessed in archival collected tumour samples in our central laboratory using 4 commercial kits for immunoistochemical (ICH) analysis (clone 22C3 pharm DX Dako Agilent, 28.8 Abcam and SP142 and SP263 Ventana Medical System). A tumor cell was considered positive if any membranous staining was found regardless of the intensity. In particular the immunostaining was scored 0 when all tumor cells were unstained (PD-L1-negative), 1+ when < 1% positive tumor cells were counted, 2+ when the percentage was between 1% and 50%,3+ when the number of stained cells was more than 50%. ORR and overall survival (OS) were correlated with PD-L1 staining. Results: Formalin-fixed paraffin-embedded (FFPE) specimens were obtained from 44 of 69 pts enrolled in the NIVES study. Twenty-two pts of 44 (50%) were considered PD-L1-negative using all the 4 commercial kits for ICH analysis, while 14 of 44 pts (31,8%) were defined PD-L1 weakly positive (positive tumor cells < 1% at least in one kit for ICH). Eight of 44 pts (18.1%) were defined PD-L1 strong positive when at least one kit for ICH scored 2+ or 3+. About the correlation between ORR and PDL1 staining in the 42 pts (2/44 pts are not evaluable for ORR), ORR was 18.2% (95% CI, 5.2% to 40.3%) in the PD-L1-negative group vs 20% (95% CI, 5.7% to 43.7%) in weakly/strongly PD-L1 positive (p = 1.00). Among the 44 pts in the intention-to-treat population with available PD-L1 status, median OS was not significantly different between pts with PD-L1 negative (20.56 months, 95% CI, 7.16 to NR) and PD-L1 positive (18.33 months, 95% CI, 6.83 to NR) (p = 0.56). Conclusions: For the first time four commercial kits for ICH analysis were used to test PD-L1 expression in pretreated mRCC pts. Data from these small sample size seem to confirm that PD-L1 in pre-treated mRCC cancer is not a predictive biomarker for selecting pts to receive NIVO-based treatment. Clinical trial information: NCT03469713.

Epidemiology and clinical course of severe acute respiratory syndrome coronavirus 2 infection in cancer patients in the Veneto Oncology Network: The Rete Oncologica Veneta covID19 study

Introduction

Coronavirus disease 2019 (COVID-19) pandemic started in Italy with clusters identified in Northern Italy. The Veneto Oncology Network (Rete Oncologica Veneta) licenced dedicated guidelines to ensure proper care minimising the risk of infection in patients with cancer. Rete Oncologica Veneta covID19 (ROVID) is a regional registry aimed at describing epidemiology and clinical course of Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection in patients with cancer.

Materials and methods

Patients with cancer diagnosis and documented SARS-CoV-2 infection are eligible. Data on cancer diagnosis, comorbidities, anticancer treatments, as well as details on SARS-CoV-2 infection (hospitalisation, treatments, fate of the infection), have been recorded. Logistic regression analysis was applied to calculate the association between clinical/laboratory variables and death from any cause.

Results

One hundred seventy patients have been enrolled. The median age at time of the SARS-CoV infection was 70 years (25–92). The most common cancer type was breast cancer (n = 40). The majority of the patients had stage IV disease. Half of the patients had two or more comorbidities. The majority of the patients (78%) presented with COVID-19 symptoms. More than 77% of the patients were hospitalized and 6% were admitted to intensive care units. Overall, 104 patients have documented resolution of the infection. Fifty-seven patients (33%) have died. In 29 cases (17%), the cause of death was directly correlated to SARS-CoV-2 infection. Factors significantly correlated with the risk of death were the following: Eastern Cooperative Oncology Group performance status (PS), age, presence of two or more comorbidities, presence of dyspnoea, COVID-19 phenotype ≥ 3, hospitalisation, intensive care unit admission, neutrophil/lymphocyte ratio and thrombocytopenia.

Conclusions

The mortality rate reported in this confirms the frailty of this population. These data reinforce the need to protect patients with cancer from SARS-CoV-2 infection.

Impact of COVID-19 outbreak on cancer immunotherapy in Italy: a survey of young oncologists

Background

The coronavirus disease 2019 (COVID-19) pandemic has overwhelmed the health systems worldwide. Data regarding the impact of COVID-19 on cancer patients (CPs) undergoing or candidate for immune checkpoint inhibitors (ICIs) are lacking. We depicted the practice and adaptations in the management of patients with solid tumors eligible or receiving ICIs during the COVID-19 pandemic, with a special focus on Campania region.

Methods

This survey (25 questions), promoted by the young section of SCITO (Società Campana di ImmunoTerapia Oncologica) Group, was circulated among Italian young oncologists practicing in regions variously affected by the pandemic: high (group 1), medium (group 2) and low (group 3) prevalence of SARS-CoV-2–positive patients. For Campania region, the physician responders were split into those working in cancer centers (CC), university hospitals (UH) and general hospitals (GH). Percentages of agreement, among High (H) versus Medium (M) and versus Low (L) group for Italy and among CC, UH and GH for Campania region, were compared by using Fisher’s exact tests for dichotomous answers and χ² test for trends relative to the questions with 3 or more options.

Results

This is the first Italian study to investigate the COVID-19 impact on cancer immunotherapy, unique in its type and very clear in the results. The COVID-19 pandemic seemed not to affect the standard practice in the prescription and delivery of ICIs in Italy. Telemedicine was widely used. There was high consensus to interrupt immunotherapy in SARS-CoV-2–positive patients and to adopt ICIs with longer schedule interval. The majority of the responders tended not to delay the start of ICIs; there were no changes in supportive treatments, but some of the physicians opted for delaying surgeries (if part of patients’ planned treatment approach). The results from responders in Campania did not differ significantly from the national ones.

Conclusion

Our study highlights the efforts of Italian oncologists to maintain high standards of care for CPs treated with ICIs, regardless the regional prevalence of COVID-19, suggesting the adoption of similar solutions. Research on patients treated with ICIs and experiencing COVID-19 will clarify the safety profile to continue the treatments, thus informing on the most appropriate clinical conducts.

Nivolumab (NIVO) in combination with stereotactic body radiotherapy (SBRT) in pretreated patients (pts) with metastatic renal cell carcinoma (mRCC): First results of phase II NIVES study

613

Background: NIVO showed an increased on OS in pre-treated mRCC. The introduction of metastasis SBRT could improve the clinical outcomes. NIVES Study evaluated the efficacy and safety of SBRT in combination with NIVO in II and III line of mRCC pts. Methods: This is a phase II, single arm, multicentre study in mRCC pts with PD after ≤2 prior anti-angiogenic therapies with measurable metastatic sites, and at least one suitable for SBRT. The pts received hypofractionated radiation in 1 lesion at dose of 10 Gy/3 fractions after 7 days from the first infusion of NIVO. NIVO is given as flat dose of 240 mg on day 1 every 14 days for 6 months, then 480 mg q4-weekly in responding pts until PD or unacceptable toxicity. The primary end point is ORR. Secondary endpoints are PFS, OS, ORR of irradiated and non-irradiated metastasis and safety profile. Results: 69 pts were enrolled from July 2017 to March 2019 in 12 Italian centers. Pt characteristics were: 79.7% clear cell histology, 82.6% males, 79.7%% IMDC intermediate/poor, median age 67 yrs (43-85), 18.8% third line, 21.7% non-nephrectomy. The most frequent sites of SBRT were lung (37.7%), lymphonodes (15.9%), bone (11.6%). At the time of this analysis, median number of NIVO doses received was 12 (1-32). The ORR was 19% (1 CR) and DCR 63.5% (no statistically difference between site of SBRT and ORR); the largest benefit in pts with clear cell histology (p=0.01). Median PFS was 4 months (95%CI, 2.8-7.1), median OS 22.1 months (95%CI, 18.1-NR). With a median follow-up of 15 months (0-25.6), 6-month and 9-month survival rates were 80.3% and 56.1% respectively. 7 pts (10.1%) discontinued treatment due to AEs; grade(G) 3-4 toxicities related to NIVO were experienced in 17 pts (24.6%). The most frequent G3-4 toxicities included diarrhea (5.8%), amylase/lipase increased (4.3%) and hypothyroidism (4.3%); no G3-4 toxicities related to SBRT. Conclusions: The NIVES Study represents the first prospective trial of NIVO and SBRT combination in pre-treated mRCC pts. At present the Study showed a high DCR and no-increase of toxicity. It is ongoing the analysis of correlation between efficacy and PD-L1 expression. Clinical trial information: NCT03469713.

Renal cell carcinoma in one year: Going inside the news of 2017 - A report of the main advances in RCC cancer research

Very interesting issues regarding RCC treatment have been raised during 2017. We analysed the main news that may potentially modified clinical practice. Conflicting data came from trials testing targeted therapies in the adjuvant setting, supporting the necessity of further investigations. One of the key goals of RCC research is focused on the first-line therapy, with particular interest focus on immunotherapy combinations. Redefine the standard of care with the aim of improving patients' survival represents an imperative need. Enhancing immunotherapy antitumor activity by combining immune checkpoint inhibitors with anti-angiogenetic therapies is a noteworthy research field, with promising results. In addiction, we analysed in the metastatic setting data about the role of cytoreductive nephrectomy and the possibility of delay the start of first-line therapy after an active surveillance period. Based on recent developments, the paper outlines future prospective of RCC research.

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

INTRODUCTION

The 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).

METHODS

Clinical-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.

RESULTS

Data 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.

CONCLUSIONS

Despite 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.

Prognostic model for advanced breast carcinoma with luminal subtype and impact of hormonal maintenance: Implications for post-progression and conditional survival

BACKGROUND

The aim of this analysis was to develop and validate a prognostic model for advanced breast cancer (ABC) with luminal subtype based on the combination of clinical, pathological and therapeutic predictors to provide a practical tool to evaluate patients' prognosis.

METHODS

Clinical and pathological data were retrospectively correlated to progression-free and overall survival (PFS/OS) using a Cox model. Significant treatment variables were adjusted with the propensity score analysis. A continuous score to identify risk classes was derived according to model ratios. The performance of the risk-class model was tested for post-progression survival (PPS) and conditional survival (CS) as well.

RESULTS

Data from 335 patients (3 institutions) were gathered (median follow-up 58 months). At multivariate analysis Ki67, Performance Status (PS) and number of metastatic sites were significant predictors for PFS, whereas Ki67, PS, brain metastases, PFS after 1st-line therapy, number of chemotherapy lines, hormonal therapy and maintenance were significant predictors for OS. The hormonal maintenance resulted to be prognostic after adjustment with propensity score analysis. A two-class model significantly differentiated low-risk and high-risk patients for 2-year PFS (31.5% and 11.0%, p < 0.0001), and 3-years OS (57.1% and 4.8%, p < 0.0001). A three-class model separated low risk, intermediate-risk, and high-risk patients for 2-year PFS (40.8%, 24.4%, and 11.0%, p < 0.0001) and 3-year OS (68.1%, 24.8%, and 4.8%, p < 0.0001). Both models equally discriminate the luminal ABC prognosis in terms of PPS and CS.

CONCLUSIONS

A risk stratification model including 'easy-to-obtain' clinical, pathological and therapeutic parameters accurately separates luminal ABC patients into different risk classes.

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

Background

The 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.

Methods

Clinical-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.

Results

At 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.

Conclusions

Despite 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 material

The online version of this article (doi:10.1186/s13046-016-0325-z) contains supplementary material, which is available to authorized users.