Background:
The standard therapy of first-line metastatic luminal hormone receptor positive, HER2 negative breast cancer, is based upon the combination of hormone therapy (HT): an aromatase inhibitor (AI) (e.g. Anastrozole or Letrozole) and a CDK 4/6 inhibitor (e.g. Palbociclib). Around 50% of patients are benefiting from this combination. However, there are no predictive markers to identify upfront the benefiting population, thus imposing to treat the whole population (100%) for the benefit of 50%.
Objectives:
Primary:
- Rate of objective clinical response for the combination of Palbociclib + AI.
- Value of quantum optics (QO) analysis on liquid biopsies to predict the upfront efficacy/resistance of the combination (Palbociclib + AI) on an individual basis.
Secondary:
- Progression-Free Survival (PFS).
- Clinical benefit (CR/PR/Stabilization>24 weeks).
- 'Objective' clinical benefit (CR/PR/Minor response (> 0% > 24 weeks).
- Modification of the individual molecular fingerprinting during treatment.
- Role of individualized molecular fingerprinting to predict early progression (>6 months).
- Role of individualized molecular fingerprinting to predict relapse in age-matched patients with an initial response.
- Safety and tolerability of the proposed regimen.
- Benchmark analysis, using Raman Spectroscopy.
Materials and Methods:
Study design:
This study is a multicenter, international, open-label phase II study. Based on inclusion/exclusion criteria, eligible pre and post-menopausal patients with newly diagnosed metastatic Luminal HER2 negative breast cancer, will be prospectively treated with a standard combination of HT (AI: Letrozole or Anastrozole) and Palbociclib. This combination will continue until progression. Treatment response will be evaluated every three months using clinical and radiological assessments (Revised RECIST guidelines).
Patients will undergo liquid biopsies (blood tests) for plasma molecular fingerprinting by the QO technology at enrolment, each month for the first three months and then every three months until relapse, along with clinical and radiological assessments every three months.
Sample size: Considering the approximately 50% response rate (RR) previously reported in the phase III PALOMA 2 trial (AI + Palbociclib) and the initial published data of QO analysis on liquid biopsies, which compared plasma molecular fingerprint profiles between a control population without breast cancer and a population with breast cancer (sensitivity: 98% / specificity: 97%), we hypothesize that the sensitivity to differentiate responders to non-responders by QO analysis on liquid biopsies (prediction of efficacy) will be 97%. Based on the above hypothesis, the sample size calculated using alpha 5% and power 95% is N=80 patients to be enrolled and treated.
Populations:
The Intent-to-Treat (ITT): all patients enrolled in the study. The safety population: all patients receiving any treatment.
Statistical Methodology:
Qualitative variables are frequency and percentage. Comparison of the proportion between responders and non-responders will use the chi-square test.
Quantitative variables: mean, standard deviation, median, maximum, and minimum. Time-dependent parameters: Kaplan-Meier method. Cox's proportional hazards regression analysis for response rates, clinical benefits, PFS.
All tests of hypotheses: one-sided. Confidence intervals of the median survival: method of Simon.
The methodology of computational analysis: "non-hierarchical deep data mining" approach with:
- Analyses with the support vector machines with a polynomial nucleus.
- Age stratifications when applicable with sub-stratifications in different age groups.