Tumori Journal Abstract Book - October 2020 - 35

B - Gastrointestinal (Colorectal) Cancers	
Background: Cardiotoxicity represents an unexpected and
potentially life-threatening toxicity in colorectal (CRC)
patients (pts) treated with fluoropyrimidines (FP). Currently,
there are no predictive factors of FP-induced cardiotoxicity
(FIC), moreover the correlation between FIC and known
cardiovascular (CV) risk factors remains controversial.
Methods: Colorectal (CRC) FP naïve pts have been enrolled
at Candiolo Cancer Institute since January 2016 to February
2020. Before the start of FP-regimen, all patients were
screened for CV comorbidities and risk factors with an optimization of the cardiological treatment. During the first 3
chemotherapy (CT) cycles, seriated electrocardiograms and
questionnaires on CV symptoms were administered, as well
as brain natriuretic peptide (BNP) levels assessment and
plasma samples storing for the validation of circulating
microRNAs (c-miRNAs) were performed. Primary objective was to determine the incidence rate of FIC in our colorectal cancer patients treated with FP. A secondary endpoint
was to determine any predictive factors for FIC.
Results: An interim analysis was conducted at 135 pts
(60.7% men, median age 66 years). We recordered CV risk
factor in 48.7% of enrolled pts. During the treatment period
20 pts (15,38%) experienced FIC: 1 acute coronary syndrome, 1 coronary vasospasm, 1 paroxysmal supraventricular tachycardia, 1 complete left bundle branch block, 3
syncope, 4 typical chest pain, 6 sudden wheezing and 3 sudden palpitations. Unfortunately, none of clinical or laboratories index was able to suggest an increased risk of FIC.
Instead, we found a higher incidence of FIC in women (p.
0.048). The onset of FIC is not related to age ( 75 years),
cardiovascular risk (high / low), type of FP (5-fluorouracil /
capecitabine) and disease status (adjuvant / metastatic). BNP
levels increased significantly more in patients treated with
infusion-based 5-FU CT (p. 0.003) in the first 3 days of therapy, however no correlation was observed with FIC.
Conclusions: Although FIC was observed in a not negligible percentage of patients, no relationship with CV risk
factors or laboratory data was found. A more precise correlation of cardiac events with c-miRNAs will be available
at completed recruitment (200 pts).

B12
Impact of Early Tumor Shrinkage
(ETS) and Depth of Response
(DoR) on the outcomes of
panitumumab(pan)-based
maintenance therapy in patients
(pts) with RAS wild-type metastatic
colorectal cancer (mCRC) and
enrolled in the Valentino study
Manca P.1, Corallo S.1, Calareso G.1, Lonardi S.2, Cremolini C.3,
Rimassa L.4, Racca P.5, Murialdo R.6, Clavarezza M.7, Zaniboni A.8,
Tomasello G.9, Adamo V.10, Antonuzzo L.11, Vaiani M.1, Cinieri S.12, Di
Bartolomeo M.1, Niger M.1, Morano F.1, De Braud F.1, Pietrantonio F.1

35
1
IRCCS Istituto Nazionale dei Tumori, Milano; 2Istituto Oncologico Veneto,
Padova; 3AOU Pisana, Pisa; 4Humanitas Cancer Center, Milano; 5AOU Città
della Salute e della Scienza, Torino; 6IRCCS AOU San Martino-IST, Genova;
7
Ente Ospedaliero Ospedali Galliera, Torino; 8Fondazione Poliambulanza,
Brescia; 9Niguarda Cancer Center, Milano; 10A.O. Papardo, Messina; 11AOU
Careggi, Firenze; 12Ospedale Antonio Perrino, Brindisi

Background: The prognostic role of ETS and DoR in
mCRC pts receiving first-line therapy is well established.
However, these parameters were not collected in maintenance trials and their impact on maintenance tx is unknown.
Methods: VALENTINO showed that maintenance tx with
pan (arm B) achieved inferior PFS compared to 5-FU/
LV+pan (arm A) after FOLFOX/pan induction in 229 pts
with RASwt mCRC. CT scans performed every 8 weeks
were centrally reviewed to determine ETS and DoR
(defined as 20% reduction of the sum of target lesions at 8
weeks and their best % reduction) and their correlation
with outcomes. Uni and multivariate cox regressions were
used to investigate factors associated with PFS and OS.
Results: Out of 196 evaluable pts, 132 (67%) had ETS and
median DoR was -44.1%. ETS frequency did not differ by
main baseline features, except for liver limited disease
(p=0.025) and in-situ primary tumors (0.004). ETS was
associated with longer mPFS (12.5 vs 8.8 m; HR=0.66
[0.48-0.91]; p=0.010) and mOS (32.5 vs 23.1 m; HR=0.60
[0.41-0.86]; p=0.006). Maximization of log-rank statistics
identified the optimal DoR cut-off as -34% for both PFS
and OS. DoR>34% was associated with longer mPFS
(13.2 vs 7.3 m; HR=0.51 [0.37-0.70]; p<0.001) and mOS
(38.4 vs 19.4 m; HR=0.44 [0.31-0.64]; p<0.001). Notably,
ETS and DoR were highly associated (p<0.001) and 89%
of pts with ETS achieved a DoR>34%. In multivariable
models, DoR was independently associated with PFS
(p=0.024) and OS (p=0.027). In the predictive analyses
(table), the PFS benefit of adding 5-FU/LV to pan was
independent from both ETS and DoR (interaction p values
ns). However, pts with no-ETS and poorer DoR treated
with maintenance pan alone achieved extremely poor PFS
compared to other subgroups. For ETS, such detrimental
effect was evident also for OS.
arm B
 

no-ETS

mPFS
mOS
 
mPFS
mOS

7.7
18.7
DoR⩽34%
6.5
18.0

arm A

arm B

arm A

ETS
9.5
24.7
8.0
19.7

11.1
33.1
DoR>34%
11.4
33.1

13.2
30.6
14.6
38.0

Conclusion: ETS and DoR are relevant prognostic factors
in an extended-RAS wt population. No-ETS and poorer
DoR are associated with unfavourable outcomes after
maintenance tx, with particular regard to pan alone.
Tumori Journal Abstract Book - October 2020

Table of Contents for the Digital Edition of Tumori Journal Abstract Book - October 2020
