Page 627 - Read Online

Peyvandi et al. J Cancer Metastasis Treat 2019;5:44 I http://dx.doi.org/10.20517/2394-4722.2019.16 Page 15 of 24

Our and other’s results indicate that chemotherapy induces an effective anti-tumor immune response upon
tumor cell treatment [214,226,227] . Clinically this implies that neo-adjuvant/pre-operative chemotherapies may
be more effective in inducing a long-lasting, protective immune response compared to classical adjuvant/
post-operative therapies, due to the larger targeted tumor mass [224,260] . Indeed, neoadjuvant chemotherapy is
+
already used in highly proliferative breast cancers (i.e., Luminal B, HER2 and TNBC) with high frequencies
of pathological complete responses (pCR). Interestingly, paclitaxel, in combination with trastuzumab,
+
induced a high rate of pCR in HER2 patients, likely due to the synergy between the immunomodulating
[221]
properties of these drugs [260] . Ladoire et al. reported that pCR to breast cancer neoadjuvant chemotherapy
+
+
was associated with the disappearance of tumor-infiltrating FOXP3 Tregs and recruitment of CD8 T cells,
consistent with the induction of an antitumor immune response by chemotherapy.
Type I IFN
In our experimental model we have shown that Type I IFN response is essential to elicit a state of
immunological breast cancer dormancy [226] . Others have shown that exogenous addition of type I IFN
boosted an insufficient response to chemotherapy in an experimental model of breast cancer and that a
type I IFN-related signature predicted clinical responses to anthracycline-based chemotherapy in breast
cancer patients [227] . We demonstrated that patients with high levels of serum IFN-β during neoadjuvant
therapy have a better outcome compared to patients with low levels [226] . This suggests that administration of
type I IFN during neo-adjuvant therapy may be effective in mounting a long-lasting immune response in
particular in those patients with low endogenous type I IFN levels. Type I IFN, in particular IFN-α has been
already tested as immunostimulatory anti-cancer agent, especially in melanoma and kidney cancers, albeit
with mild results, in part also due to the need of repeated administrations and its intrinsic toxicity [261-263] .
Alternatively, less toxic inducers of Type I IFN response, such as TLR-ligands of STING stimulators may be
considered [227,264] .

Check point inhibitors
A complementary strategy to IFN administration could be the use of check point inhibitors, in particular
anti-PD-1/PD-L1 antibodies to relieve tumor-induced immunosuppression [265] . In breast cancer,
immunotherapy is being explored, in particular in patients with tumors expressing PD-L1 and infiltrated
with lymphocytes [266] . Potential response to PD-1 or PD-L1 inhibitors was demonstrated in metastatic
+
TNBC [267] and HER2 breast cancers [268] . However, because the number of potential neoantigens available
for immune response in most breast cancers, responses are modest compared with other cancers such as
lung and melanoma, the use of check-point inhibitors may require combination with other therapies [269] .
Therefore, combination with neo-adjuvant chemotherapy (causing the release of tumor antigens), anti-
angiogenesis therapies (suppressing inhibitory cues) [182] or Type I IFN (acting immunostimulating) [270] may
be more effective and should be explored.

Biomarkers of dormancy
There are currently no specific non-invasive biomarkers to monitor breast cancer dormancy of clinical
utility [271] . Such markers would allow personalized follow up and accelerate therapeutic decisions in case of
-
+
evidence of disease progression. CD44 /CD24 CTC subsets along with combinatorial expression of uPAR
and b1 integrin, as well as proliferation and apoptosis markers in CTC of early breast cancer patients, have
been explored for potential use as biomarker of dormancy or aggressiveness [272,273] . Genomic analysis (i.e.,
SNP/CNV) of circulating ctDNA was shown to potentially identify breast cancer patients with dormant/
minimal residual disease [274] . Also, serum inflammatory markers might serve as biomarkers of relapse in
disease-free patients, as inflammation is associated with escape from dormancy but will likely be unspecific
and of limited sensitivity [275] . Serum IFN-β levels were associated with longer DMFS (as a surrogate of
dormancy) in our model of chemotherapy-induced dormancy and during neoadjuvant therapy in patients
with favorable outcome [226] . Also, we observed a shift in peripheral blood leucocyte populations in our