Vienna, Austria

ESTRO 2023

Session Item

Saturday
May 13
10:30 - 11:30
Stolz 1
Breast
CAROLINA DE LA PINTA, Spain;
Liesbeth Boersma, The Netherlands
Mini-Oral
Clinical
Predictive role of microvascular health status for RT acute toxicities in breast cancer patients.
Eliana La Rocca, Italy
MO-0140

Abstract

Predictive role of microvascular health status for RT acute toxicities in breast cancer patients.
Authors:

Eliana La Rocca1, Francesco Pisani2, Fabio Badenchini2, Luca Possenti2, Alessandro Cicchetti2, Eliana Gioscio2, Carlotta Giandini3,4, Riccardo Ray Colciago3,5, Maria Carmen De Santis3, Riccardo Valdagni3,2,4, Tiziana Rancati2

1Fondazione IRCCS Istituto Nazionale Tumori di Milano , Department of Radiation Oncology, Milano, Italy; 2Fondazione IRCCS Istituto Nazionale Tumori di Milano, Prostate Cancer Program, Milano, Italy; 3Fondazione IRCCS Istituto Nazionale Tumori di Milano, Department of Radiation Oncology, Milano, Italy; 4University Of Milan, Department of Oncology and Hemato-Oncology, Milano, Italy; 5University of Milan Bicocca, School of Medicine and Surgery, Milano, Italy

Show Affiliations
Purpose or Objective

To study the role of healthy/unhealthy microcirculation in predicting acute toxicity after adjuvant  RT for breast cancer (BC).

Material and Methods

We enrolled BC patients (pts) treated with hypofractionated RT after conservative surgery. The total dose delivered was 42.4 Gy plus 10 Gy for the boost on the tumour bed. We assessed each patient's baseline sublingual microvasculature (MV) health status on the first day of RT. A sidestream dark field camera coupled to the GlycoCheck™ software (Microvascular Health Solutions Inc, USA) was placed under the tongue by the pt herself. The system records videos showing the live movement of red blood cells (RBCs) in the microvessels and automatically computes functional parameters. Specifically, the following parameters were calculated (figure 1):

-Perfused boundary region (PBR, in µm), an estimate of the dynamic lateral movement of RBCs into the permeable part of the endothelial glycocalyx layer. Higher PBR values result from damaged glycocalyx (i.e. indicate an impaired microcirculation)

-Total density of capillaries

-Blood Flow in the analysed area (103 µm3/s/mm2)

-MV Health Score (MVHS™), higher values indicate healthier MV. MVHS is computed by weighting the information from the capillary density, the PBR, the blood flow and the recruitment capacity. Calculating MVHS requires at least 8 videos (i.e. needs more pts compliance), while the 3 functional parameters can be evaluated even with a single video.

One radiation oncologist scored the toxicities before and at the end of RT, according to CTCAE v.4.0.

We used univariate and multivariate logistic regression to assess the association between MV functional parameters and toxicity.


Results

We evaluated 63 pts. The median age at RT start was 55 yrs (range 40-75 yrs); 45 (71%) had an evaluation of the MVHS. After RT, 13 pts experienced G1+ oedema  (20.6%) and 23 G2+ erythema (36.5%).

The MVHS was associated with erythema, a healthy MV protects from this toxicity (OR=0.49, p=0.007, AUC=0.81). A multivariate model including the separated functional parameters was also associated with erythema (figure 2, AUC=0.72): higher PBR is a risk (p=0.025, OR=1.04), higher capillary density protects (p=0.05, OR=0.99), while a higher blood flow is a risk (p=0.04, OR=1.006).

Oedema associated with flow (risk factor, p=0.06, OR=1.0023) and PBR (risk factor, p=0.06, OR=1.03), AUC=0.74.






Conclusion

The MV health status predicts acute toxicity in BC pts: in our study population, an impaired MV was associated with higher susceptibility to RT.

The role of DNA damage repair mechanisms in microcirculation diseases was already known in the literature. These data suggest an interaction of compromised MV with repair mechanisms needed to repair RT damage. This kind of systemic functional information derived by a sublingual microscope could be exploited in different cancer districts, boosting the personalisation of predictive models and their tailoring to the single patient functional status.
This study was funded by AIRC Investigator Grant, no. IG21479.