Page 91 - Read Online

Page 6 of 16 D’Amico et al. J Cancer Metastasis Treat 2021;7:3 I http://dx.doi.org/10.20517/2394-4722.2020.93

Taking in consideration the clinical evidence produced in the following years, partly due to the deployment
of new technologies, here we address the potential prognostic or predictive role of ctDNA detection before
any primary treatment (neoadjuvant, surgical, adjuvant) and the possible implications.

Limited and contrasting evidence is available in support of ctDNA as a predictive biomarker of response
to neoadjuvant treatment. In this regard, in a translational ctDNA sub-study of the Neoadjuvant Lapatinib
[25]
and/or Trastuzumab Treatment Optimisation trial (NeoALTTO) , among 69 patients with human
epidermal growth factor receptor 2 (HER2) positive+ breast cancer bearing the mutation of TP53 or
PIK3CA gene in the primary tumor, the detection of the corresponding alteration in plasma before
neoadjuvant treatment was associated with failure to achieve a pathological complete response (pCR).
Interestingly, neither detection during week 2 neoadjuvant treatment nor prior surgery were associated with
pCR. In this study, ctDNA detection before any treatment was not associated with the time to recurrence
(TTR), although the analysis was underpowered to assess the correlation with both TTR and failing to
achieve a pCR. Similar results were reported from a prospective study with 55 early breast cancer patients
[26]
with different subtypes at the diagnosis , which demonstrated that ctDNA detection with personalized
digital polymerase chain reaction (dPCR) assays of somatic mutations is feasible, but fail to have a
prognostic value if detected before neoadjuvant treatment. Conversely, in the same timepoint, a significant
association between ctDNA detection and relapse was, instead, found in a bigger prospective cohort of 101
early-stage breast cancer patients [hazard ratio (HR), 5.8; 95% confidence interval (CI): 1.2-27.1; P = 0.01],
[27]
irrespective of hormone receptor and HER2 status . Due to the conflicting results obtained from these
investigations, the prognostic role of ctDNA detection before any treatment remains a matter of debate.

Several other studies addressed the prognostic potential of ctDNA detection before surgery. For instance,
[28]
despite the modest number of patients recruited on the BRE09-146 phase II clinical trial, Chen et al.
were retrospectively capable of identifying a group of triple negative breast cancer (TNBC) patients with
residual disease after neoadjuvant treatment with detected ctDNA and a very poor prognosis. Although
with high specificity (100%) but very low sensibility (31%), the patients identified by ctDNA detection had
a significantly inferior disease-free survival (DFS) if compared with relapsed patients and negative ctDNA (P
< 0.0001, median DFS: 4.6 months. vs. Not Reached (NR); HR = 12.6, 95%CI: 3.06-52.2). Considering that
only 1 ml of blood was dedicated to this analysis, this study represents the “worst-case scenario” in which
the detection of ctDNA identifies an exceptionally high-risk patient population. However, the previous
study is not the sole addressing and supporting the potential clinical role of ctDNA in this timepoint.
Indeed, to the present day, one of the strongest pieces of evidence is the preplanned analyses from the
phase II trial BRE12-158 presented by Radovich et al. . DFS, distant DFS (DDFS), and OS at 24 months of
[29]
196 patients with early-stage TNBC, who had residual disease after neoadjuvant treatment, were analyzed.
Detection of ctDNA before surgery was significantly associated with an inferior DFS (HR = 2.67, 95%CI:
1.28-87 5.57; P = 0.009), DDFS (median DDFS: 32.5 months vs. Not Reached; HR = 2.99, 95%CI: 1.38-6.48;
P = 0.006) and OS (HR = 4.16, 95%CI: 1.66-10.42; P = 0.002). Notably, the distant DFS probability was 56%
in ctDNA-positive patients as compared to 81% in ctDNA-negative patients. Taken together, these results
indicate that the detection of ctDNA in early-stage TNBC after neoadjuvant treatment is an independent
predictor of disease recurrence.

Leveraging the lessons we have learned from these trials, it appears legitimate to further investigate the
potential of ctDNA, among the different breast cancer molecular subtypes, to select high risk patients that
could benefit from more demanding adjuvant treatment, sparing, in the meanwhile, unnecessary toxicity
to those at lower risk of recurrence. Similar strategies are being adopted and investigated in trials enrolling
stage II colon cancer. (“The DYNAMIC study- Circulating tumor DNA analysis informing adjuvant
chemotherapy in Stage II Colon Cancer”, registration number: ACTRN126150003815830).