Page 162 - Read Online

Battaglin et al. J Cancer Metastasis Treat 2018;4:12 I http://dx.doi.org/10.20517/2394-4722.2018.04 Page 7 of 25

role in tumor microenvironment immuno-modulation and anti-angiogenic treatment has been proposed as
[93]
an effective modality to potentiate immunotherapy . No definitive evidence is available on the prognostic
role of MSI-H in mCRC; recent data suggest no statistically significant difference in OS between MSI-H and
MSS mCRCs, although a trend toward a worse OS has been reported for MSI-H . Some studies suggest the
[94]
correlation with BRAF mutational status as a potential confounding factor affecting the estimation of MSI-H
impact on survival in mCRC . However, the prognostic role of BRAF in these tumors is still object of debate
[95]
and in a recent analysis BRAF V600E mutation was not associated with a worse survival in MSI-H CRC .
[80]
Additionally, a possible negative prognostic effect of immune checkpoint expression in MSI-H CRCs have
been recently reported, which seems to be able to counterbalance the positive effect of tumor-infiltrating
cytotoxic T-cell lymphocytes in these tumors .
[96]

MSI assessment has lately gained a prominent role in the metastatic setting due to the recent groundbreaking
success of immunotherapy with checkpoint inhibitors in MMR-D mCRCs which has opened a new era
in the treatment of MSI-H tumors. In the phase II KEYNOTE 016 trial, pembrolizumab demonstrate its
activity in 28 MSI-H mCRC patients with refractory disease, significantly improving response rate (RR),
disease control rate (DCR), median PFS and OS compared to MSS patients (RR: 50% vs. 0% and DCR 89% vs.
16%, respectively; HR for PFS = 0.135, P < 0.001, HR for OS = 0.247, P = 0.001) [97,98] . The combination of ipilimumab
(an anti-CTLA4) and nivolumab (an anti-PD1), under investigation in the phase II CHEKMATE142 trial,
showed as well significant results with a recently reported RR of 31.1% (95% CI, 20.8-42.9) in patients receiving
nivolumab (n = 74) and 55% (95% CI, 45.2-63.8) in those receiving ipilimumab plus nivolumab (n = 119),
and remarkable 12 months PFS rate and 12 months survival rate (50% and 73% respectively, for nivolumab
monotherapy; 71% and 85% respectively, for nivolumab plus ipilimumab) [99,100] . Responses were irrespective
of tumor RAS and BRAF mutational status, immune cell PD-L1 expression or clinical history of LS. Notably,
both pembrolizumab and ipilimumab/nivolumab showed a trend towards a plateau in the tail of patients’
survival curves, suggesting the possibility of long term responders similar to the previous experience with
immunotherapy in melanoma. Following these striking results, FDA approval was granted for the use of
checkpoint inhibitors pembrolizumab (Keytruda®, Merck & Co., Inc.) and nivolumab (Opdivo®, Bristol-
[101]
Myers Squibb) in the treatment of MSI-H or MMR-D mCRC.
[100]
Despite the clinical success of anti-CTLA4 and PD-L1/PD-1 inhibitors, however, only a subset of selected
patients exhibits durable responses, suggesting that a broader view of cancer immunity is required. A
complex set of dynamic tumor, host and environmental factors modulate the strength and timing of immune
anticancer response, and several key immunoregulatory pathways have been identified and involved in the
definition of an immune signature to predict responses to immunotherapy [102-105] . Alongside the ongoing
extensive effort to identify additional predictive biomarkers [106,107] , understanding the mechanisms limiting
immunotherapy efficacy, both in terms of innate and acquired resistance, represents a challenge which needs
to be addressed in order to improve treatment outcomes and develop new actionable strategies [108-110] .

Dihydropyrimidine dehydrogenase
Fluoropyrimidine analog 5-FU and its pro-drug capecitabine represent the backbone of chemotherapy
treatment for colorectal cancer . The mechanism of action of these drugs is based on thymidylate
[10]
synthase (TYMS) inhibition through the formation of a ternary complex between the active metabolite
5-fluoro-2-deoxyuridine-5-monophosphate (5-FdUMP), TYMS and 5,10-methylentetrahydrofolate, leading
[111]
to the suppression of DNA synthesis . The rate-limiting enzyme for 5-FU catabolism is the enzyme
dihydropyrimidine dehydrogenase (DPD), responsible for the inactivation of more than 80% of the
administered dose of 5-FU .
[112]

Up to one-third of patients treated with these agents experience severe (and in 0.5%-1% of cases lethal)
toxicities including myelosuppression, mucositis and diarrhea . Functional DPD gene (DPYD) variants
[113]