The TNM staging system for colorectal cancer has been the primary means of prognostication for patients since its inception. In colon cancer, although surgery remains the primary treatment, adjuvant chemotherapy is administered in high-risk patients. The stratification of patients to direct such treatment decisions is currently based on the TNM system of staging.
A similar approach is adopted to stratify patients with rectal cancer. Those with early rectal cancer (T0–2, N0) will undergo immediate surgery, whereas those with locally advanced rectal cancer (LARC; T3–4 and/or N+) will receive neoadjuvant chemoradiotherapy (CRT) followed by total mesorectal excision surgery.1 , 2 For the cohort receiving CRT, there is a known spectrum of response,3 , 4 which includes a proportion of patients who will achieve a pathologic complete response (pCR), defined as no tumor remaining in the irradiated tumor bed. This has spurned an increasing number of trials assessing the safety of a watch-and-wait approach to avoid surgery for patients with a clinical complete response, acting as a surrogate marker for a pCR.5–7 However, patients managed by this new watch-and-wait algorithm have experienced tumor regrowth rates of up to 38%,8 given the lack of a high-fidelity tool that can reliably predict pCR.9
The TNM staging system has been challenged over recent years by a novel prognostic marker that demonstrated improved patient stratification for long-term outcomes. The seminal article by Galon et al10 in 2006 successfully characterized different subsets of tumor-infiltrating lymphocytes (TILs) that predict patient prognosis. This was undertaken by way of gene expression profiling in a small cohort, followed by validation with immunohistochemistry on colon cancer tissue in a larger cohort. This identified that the density and location (center of tumor and invasive margin (IM)) of CD3+ (generalized), CD8+ (cytotoxic), and CD45RO+ (memory) T cells are collectively a better predictor of patient survival than the current TNM staging system. This was additionally translated to rectal cancer, showing high predictive correlation not only with long-term survival but also with tumor regression grade (TRG).2
Subsequently, significant interest has been garnered in this field, marked by an increasing number of publications validating TILs and assessing different immune subsets and their impact on prognosis.11–13 Despite the increasing research interest, the use of TILs as a prognostic marker has yet to be translated to the clinical setting. Furthermore, some results are discordant across articles assessing similar T-cell subsets, likely because of the small patient numbers in these studies.11–13 Before the prognostic and predictive power of colorectal cancer TILs can be harnessed as part of a clinical decision-making tool, the significance of each TIL subset must be confirmed. This will shift the research focus to a limited number of TIL subsets that are most relevant to colorectal cancer. The aim of this study was to investigate the significance of each T-cell subset in primary and metastatic colorectal cancer.
MATERIALS AND METHODS
A systematic search was undertaken through PubMed and Embase from January 1996 to December 2016. The key words used in combination were colon, rectum, metastatic, neoplasm, immunology, tumor-infiltrating lymphocytes, cytotoxic, helper, regulatory cells, CD8, CD4, FoxP3, neoadjuvant therapy, radiotherapy, rectal neoplasm, and pathological response. The search was limited to human studies, and the English literature with the strategy displayed in Figure 1.
Two independent reviewers (J.C.K. and G.R.G.) assessed 373 titles and abstracts. Of these, 106 were selected for full-text review. Studies were included if immunohistochemistry was performed for the assessed T-cell subsets, with reporting of short- and long-term outcomes including disease-free survival (DFS), overall survival (OS), and TRG. The exclusion criteria included studies investigating the prognostic impact of TILs in microsatellite instability–high colorectal cancer, studies with <30 patients, studies with insufficient outcome data, and studies using only hematoxylin and eosin to identify TILs.
Three studies were randomly selected for initial review to develop a standardized pro forma for data collection. Two reviewers (J.C.K. and G.R.G.) independently evaluated and collated data from each article, with discordance addressed by a third reviewer (A.G.H.). Data collected included study characteristics, recruitment date, single or multicenter study, number of investigators, blinded or not to outcome data, multivariate analysis adjustments, patient characteristics, age, sex, assessment of TIL density, tumor stage, patient outcome, and extent of follow-up.
The primary outcome measures were DFS and OS in high versus low TIL colorectal cancers. The secondary outcome measure was TRG after neoadjuvant CRT for high versus low TIL rectal cancer.
In each study examining primary or metastatic colorectal cancer, time-to-event outcomes were extracted as an HR with the accompanying 95% CI. The estimated HR compared high-density versus low-density TILs, with an HR <1 associated with good outcome and >1 signifying poor outcome. Subgroup analyses were performed based on T-cell subset, density, location, and the associated long-term outcomes.
T-cell location was characterized as follows: 1) IM, defined as TILs abutting the IM of the tumor; 2) intraepithelial, with TILs identified between tumor cells or nests and 3) stromal, with TILs identified in an area clearly separated from the luminal border and invasive front. OS was defined as the time from the date of diagnosis of colorectal cancer to the date of death irrespective of cause, with DFS defined as the date of diagnosis to the date of cancer recurrence.
Pooled data analysis was performed using the random-effects model given the significant heterogeneity across the included studies.14 This model is an extension of the Bayesian meta-analysis suggested by Jackson et al,15 with the addition of a model fit using importance sampling and a likelihood-based approach.14 I2 statistic was derived to assess interstudy heterogeneity, and an I2 > 60% was considered to demonstrate substantial heterogeneity.
With the LARC articles, there was substantial heterogeneity among T-cell subsets, neoadjuvant treatment, tumor regression grading, and reporting of short-term and long-term outcomes. Moreover, not all of the studies reported on the proportion of pCR in relation to TIL density. The quality of each study was assessed using the Ottawa–Newcastle scale. All of the statistical analyses were performed on IBM SPSS version 22 (IBM Corp, Armonk, NY), and a p value of <0.05 was considered significant.
Study and Patient Characteristics
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow chart including the criteria for selection of studies is demonstrated in Figure 1. There were 25 studies included in the final analysis. This consisted of 15 studies10–13 , 16–26 (4719 patients) of primary colorectal cancer and 7 studies1 , 2 , 27–31 of LARC (727 patients) investigating TIL correlation with TRG. An additional 3 studies32–34 (418 patients) examined metastatic (liver and/or lung) colorectal cancer. The majority were retrospective observational studies derived from prospectively maintained databases. The quality of each study is presented in Table S1 (Supplemental Digital Content 1, http://links.lww.com/DCR/A903), with all of the studies obtaining a score of >6 on the Ottawa–Newcastle scale. Patient, tumor, and study characteristics are displayed in Table 1.
Pooled Analysis for T-Cell Subsets in Primary Colorectal Cancer
Three T-cell subsets were identified as significant predictors of long-term outcome irrespective of location. High CD8+, FoxP3+, and CD45RO+ densities were associated with improved DFS and OS (see Table 2) with pooled estimated OS HRs of 0.83, 0.20, and 0.63 (all p values < 0.001), respectively. However, significant heterogeneity was identified in CD8+ pooled studies (8 studies; I2 test = 68.75%) for OS.
T-Cell Locations in Primary Colorectal Cancer
This was assessed to determine the prognostic impact of the location of TILs and their respective subsets. There was a single study investigating the effect of CD3+ and CD8+ TILs located at the IM of the tumor on DFS.12 This demonstrated that both T-cell subsets correlated with improved DFS and OS, with reported HRs of 0.66 (p = 0.005) and 0.76 (p = 0.046) respectively. When considering an intraepithelial location, only high intraepithelial, only high densities of CD8+ (Fig. 2) were associated with improved DFS (pooled estimated HRs are shown in Table 2).
In examining OS, all T-cell subsets (except for CD3+ in intraepithelial region) in the IM, intraepithelial, and stromal regions were significantly correlated with improved outcome, however only 1 study by Richards et al35 investigated their relevance within the IM. Moderate heterogeneity was identified in the studies examining CD3+ cells within the intraepithelial compartment (4 studies; I2 test = 71.73%) and CD8+ cells in unspecified or combined locations (4 studies; I2 test = 82.29%).
T-Cell Subsets in Metastatic Colorectal Cancer
All of the included studies for metastatic colorectal cancer reported OS as the only long-term outcome. In all 3 of the studies, there was no association identified between CD4+ and CD8+ T-cell density with OS.
T-Cell Subsets Predicting pCR
Of the 7 studies examined, 4 performed immunohistochemistry on pre-neoadjuvant and post-neoadjuvant CRT specimens,1 , 27 , 29 , 30 2 on pre-neoadjuvant CRT biopsy samples,2 , 31 and 1 on post-neoadjuvant CRT specimens only.28 A consistent correlation was seen between the density of each T-cell subset with TRG, DFS, and OS (see Table 3). Three studies showed significantly higher levels of CD3+ and CD8+ TILs after neoadjuvant CRT,1 , 29 , 30 whereas the FoxP3+ density was reduced after treatment, all of which were significantly associated with an improved TRG.1 , 28
Interest and research into the role of TILs has expanded rapidly over recent decades. This has enhanced our knowledge and understanding of tumor biology and its interaction with the immune system within the tumor microenvironment. This analysis affirms the prognostic importance of all relevant T-cell subsets (CD3+, CD8+, CD4+, CD45RO+, and FoxP3+) on DFS and OS in primary and metastatic colorectal cancer, as well as on TRG in LARC, to further delineate their clinical importance.
The concept of cancer immune surveillance was first described by William Coley36 in the 1890s, after his observations of tumor response to an injection with bacteria. Nevertheless, this concept did not gain significant traction until 2006, when Galon et al10 successfully demonstrated the significance of CD3+ T cells and the cytotoxic (CD8+) and memory (CD45RO+) subsets in differentiating good from poor prognosis colon cancer. This seminal work subsequently led to the concept of an Immunoscore, which serves as a robust classification system based on the immune infiltrate at the center and IM of colon cancer specimens. The Immunoscore has been validated as superior to the American Joint Committee of Cancer TNM classification in stratifying patient long-term outcomes.37
Since then, an international validation of this system involving 2681 patients was published recently.38 This study successfully demonstrated a high level of reproducibility between observers and centers, and that the Immunoscore is an independent predictor of time to recurrence when compared with other known prognostic factors (age, T stage, N stage, and microsatellite instability status).38 Therefore, this adds additional impetus to incorporate the immune contexture of each patient’s colorectal cancer into their staging as a TNM-immune classification.
Because of the numerous immune cell subsets within the tumor microenvironment and the inconsistent results reported in multiple studies, it was challenging to determine which T-cell subsets were significant in predicting patient DFS and OS. Regulatory (FoxP3+) T-cells are known to suppress antitumor activity, leading to poor patient prognosis. However, there were 4 studies that revealed that a high FoxP3+ density was associated with improved DFS and OS,13 , 18–20 contradicting the theorized biological function of this subset.16 , 17 , 22
In our analysis, a high FoxP3+ density was associated with improved DFS and OS in both intraepithelial and stromal regions. This discrepancy in results could be explained by the finding of FoxP3+ in normal cells or other tumor types.39 , 40 Consequently, relying on a single FoxP3+ stain is not a robust method for identifying T-regulatory cells, with a minimum of 2 markers required for accurate phenotyping (CD4+ and FoxP3+ cells).41 This highlights the importance of exploring other methods to identify T-regulatory cells, including flow cytometry or multiplex fluorescence immunohistochemistry, given their ability to costain for ≥2 markers within a single T-cell.42
A high density of the other T-cell subsets (CD3+, CD8+, and CD45RO+) correlated with improved outcome, consistent with that expected given their biological function. Cytotoxic (CD8+) T-cells are the key subset of the immune system responsible for tumor killing. The presence of CD45RO+ on T-cells infers a memory phenotype, indicating that the immune system has successfully recognized and been appropriately primed against a tumor antigen.43 From a clinical perspective, improving the accuracy of predicting treatment response and patient outcomes allows for the stratification of patients, which in turn informs treatment decision making. A worldwide taskforce has been established on the classification of tumors using the Immunoscore as a biomarker. Our meta-analysis affirms the prognostic use of immune cells in primary colorectal cancer.44
The relationship of microsatellite instability high (MSI-H) colorectal cancer and TILs has been well established,45 with an understanding that these tumors are highly immunogenic because of a hypermutated state, secondary to the loss of DNA mismatch repair activity.46 These tumors have a high lymphocytic infiltration and are associated with a better prognosis compared with microsatellite stable tumors.47 Given this established relationship, our study excluded MSI-H tumors.
Over recent time there has been the advent of a watch-and-wait approach to the management of patients with LARC who achieve a complete clinical response postneoadjuvant CRT. However, a recent propensity score–matched multicenter cohort analysis by the Oncological Outcomes After Clinical Complete Response in Patients With Rectal Cancer group found that 38% of patients in the watch-and-wait cohort had local tumor regrowth, of which 88% were salvageable by surgery. Despite no difference in 3-year OS, there is a need for a predictive test that can more accurately identify patients who have achieved a pCR to bring personalized medicine to this field. This will reduce the morbidity associated with tumor regrowth and the anxiety of having intensive clinical and radiologic follow-up over 5 years.8
This has led to an expanded research effort to identify novel biomarkers of response, exemplified by the Immunoscore.2 Other studies have successfully validated the use of T-cell subsets as predictive biomarkers of response. This has included Lim et al,30 Teng et al,29 and Shinto et al,1 who identified that an increase in CD8+ T-cells after neoadjuvant CRT is significantly associated with pCR. Furthermore, the predictive accuracy for pCR was increased when the ratio of CD8+ to FoxP3+ was higher,1 adding weight to the premise that T-regulatory cells have an immunosuppressive effect on antitumor immunity. Similar findings were identified by McCoy et al,28 where low FoxP3 correlated with an improved TRG, in addition to DFS and OS. This is consistent with a review by Haikerwal et al,48 which concluded that the immunosuppressive effect of T-regulatory cells is a delayed event, consistent with their role in regulating the resolution of inflammation. A larger prospective validation cohort is required before the T-cell subsets can be used as a clinical decision marker.
In metastatic colorectal cancer, an increased density of T-cells was not correlated with OS, however, MSI-H studies were excluded from this analysis. The potential clinical relevance in the metastatic cohort of patients, rests on the advent of immunotherapy, including checkpoint inhibition (PD-1/PD-L1 blockade). These new therapies have demonstrated effectiveness across multiple tumor types, including melanoma, lung cancer, and renal cell cancer.49 , 50 However, they have not demonstrated similar efficacy in colorectal cancer51 , 52 until recently, when Le et al46 confirmed that MSI-H tumors had a significant response with PD-1 blockade. This has opened another therapeutic avenue for select patients with MSI-H metastatic colorectal cancer.
For patients without a significant antitumor immune response, advances in the field of cancer vaccination have led to them being used in an effort to recruit cytotoxic T-cells that recognize specific tumor markers.53 One marker of particular interest in colorectal cancer is mutant KRAS, which occurs in 50% of all colorectal cancer. A phase I/II clinical trial by Gjertsen et al54 used a KRAS mutant peptide vaccine in combination with granulocyte-macrophage colony-stimulating factor for patients with advanced pancreatic cancer. This led to a RAS mutation–specific T-cell immune response in 58% of patients, with an associated improved OS.54 These findings have led to the establishment of an open-label phase Ib clinical trial assessing the safety and efficacy of KRAS cancer vaccination with a checkpoint inhibitor in recurrent rectal cancer (NCT02933944, clinicaltrials.gov).
The strength of this study relies on the inclusion of the key subsets of T-cells involved in an antitumor (CD3+, CD8+, CD45RO+ cells) or protumor immune response (FoxP3+ cells). Furthermore, to the best of our knowledge, this is the first meta-analysis investigating the prognostic impact of TILs in primary and metastatic colorectal cancer, as well as the predictive accuracy of the TIL infiltrate in determining response in LARC. However, these results should be interpreted with caution, because there are several limitations.
First, all of the included studies are retrospective in nature, and there is significant heterogeneity between many studies. Second, the subset analyses were based on a small number of studies. Third, the cutoff values were different for each study, with improved guidance on measuring and classifying tumors now offered by the recent publication of Pages et al from the Immunoscore Taskforce.38 Finally, the meta-analysis is subject to publication bias, because unpublished negative findings are not accessible.
There is mounting evidence that the immune infiltrate plays an important role in predicting prognosis in colorectal cancer and tumor regression after neoadjuvant CRT in LARC. There appears to be an intricate interplay of antitumorigenic and protumorigenic immune subsets, and researchers are now in a unique position to build on the work undertaken in parallel tumor streams. This will facilitate our understanding of the colorectal cancer microenvironment and how this can be modulated to improve outcomes. It will also guide the use of immune markers in developing a robust predictive tool for the stratification of patients to inform treatment decision-making, particularly with regard to the use of immunotherapy. This will provide another avenue of treatment for those patients who at present receive little benefit from conventional treatment.
1. Shinto E, Hase K, Hashiguchi Y, et al. CD8+ and FOXP3+ tumor-infiltrating T cells before and after chemoradiotherapy for rectal cancer. Ann Surg Oncol. 2014;21(suppl 3):S414–S421.
2. Anitei MG, Zeitoun G, Mlecnik B, et al. Prognostic and predictive values of the immunoscore in patients with rectal cancer. Clin Cancer Res. 2014;20:1891–1899.
3. Damin DC, Lazzaron AR. Evolving treatment strategies for colorectal cancer
: a critical review of current therapeutic options. World J Gastroenterol. 2014;20:877–887.
4. Agostini M, Crotti S, Bedin C, et al. Predictive response biomarkers in rectal cancer neoadjuvant treatment. Front Biosci (Schol Ed). 2014;6:110–119.
5. Habr-Gama A, Perez RO, Nadalin W, et al. Operative versus nonoperative treatment for stage 0 distal rectal cancer following chemoradiation therapy: long-term results. Ann Surg. 2004;240:711–717.
6. Maas M, Beets-Tan RG, Lambregts DM, et al. Wait-and-see policy for clinical complete responders after chemoradiation for rectal cancer. J Clin Oncol. 2011;29:4633–4640.
7. Renehan AG, Malcomson L, Emsley R, et al. Watch-and-wait approach versus surgical resection after chemoradiotherapy for patients with rectal cancer (the OnCoRe project): a propensity-score matched cohort analysis. Lancet Oncol. 2016;17:174–183.
8. Kong JC, Guerra GR, Warrier SK, Ramsay RG, Heriot AG. Outcome and salvage surgery following “watch and wait” for rectal cancer after neoadjuvant therapy: a systematic review. Dis Colon Rectum. 2017;60:335–345.
9. Ryan JE, Warrier SK, Lynch AC, Heriot AG. Assessing pathological complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer: a systematic review. Colorectal Dis. 2015;17:849–861.
10. Galon J, Costes A, Sanchez-Cabo F, et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science. 2006;313:1960–1964.
11. Deschoolmeester V, Baay M, Van Marck E, et al. Tumor infiltrating lymphocytes: an intriguing player in the survival of colorectal cancer
patients. BMC Immunol. 2010;11:19.
12. Flaherty DC, Lavotshkin S, Jalas JR, et al. Prognostic utility of immunoprofiling in colon cancer: results from a prospective, multicenter nodal ultrastaging trial. J Am Coll Surg. 2016;223:134–140.
13. Salama P, Phillips M, Grieu F, et al. Tumor-infiltrating FOXP3+ T regulatory cells show strong prognostic significance in colorectal cancer
. J Clin Oncol. 2009;27:186–192.
14. Law M, Jackson D, Turner R, Rhodes K, Viechtbauer W. Two new methods to fit models for network meta-analysis with random inconsistency effects. BMC Med Res Methodol. 2016;16:87.
15. Jackson D, Barrett JK, Rice S, White IR, Higgins JP. A design-by-treatment interaction model for network meta-analysis with random inconsistency effects. Stat Med. 2014;33:3639–3654.
16. Ling A, Edin S, Wikberg ML, Öberg Å, Palmqvist R. The intratumoural subsite and relation of CD8(+) and FOXP3(+) T lymphocytes in colorectal cancer
provide important prognostic clues. Br J Cancer. 2014;110:2551–2559.
17. Richards CH, Roxburgh CS, Powell AG, Foulis AK, Horgan PG, McMillan DC. The clinical utility of the local inflammatory response in colorectal cancer
. Eur J Cancer. 2014;50:309–319.
18. Yoon HH, Orrock JM, Foster NR, Sargent DJ, Smyrk TC, Sinicrope FA. Prognostic impact of FoxP3+ regulatory T cells in relation to CD8+ T lymphocyte density in human colon carcinomas. PLoS One. 2012;7:e42274.
19. Frey DM, Droeser RA, Viehl CT, et al. High frequency of tumor-infiltrating FOXP3(+) regulatory T cells predicts improved survival in mismatch repair-proficient colorectal cancer
patients. Int J Cancer. 2010;126:2635–2643.
20. Lee WS, Park S, Lee WY, Yun SH, Chun HK. Clinical impact of tumor-infiltrating lymphocytes
for survival in stage II colon cancer. Cancer. 2010;116:5188–5199.
21. Sinicrope FA, Rego RL, Ansell SM, Knutson KL, Foster NR, Sargent DJ. Intraepithelial effector (CD3+)/regulatory (FoxP3+) T-cell ratio predicts a clinical outcome of human colon carcinoma. Gastroenterology. 2009;137:1270–1279.
22. Suzuki H, Chikazawa N, Tasaka T, et al. Intratumoral CD8(+) T/FOXP3 (+) cell ratio is a predictive marker for survival in patients with colorectal cancer
. Cancer Immunol Immunother. 2010;59:653–661.
23. Chiba T, Ohtani H, Mizoi T, et al. Intraepithelial CD8+ T-cell-count becomes a prognostic factor after a longer follow-up period in human colorectal carcinoma: possible association with suppression of micrometastasis. Br J Cancer. 2004;91:1711–1717.
24. Menon AG, Janssen-van Rhijn CM, Morreau H, et al. Immune system and prognosis
in colorectal cancer
: a detailed immunohistochemical analysis. Lab Invest. 2004;84:493–501.
25. Prall F, Dührkop T, Weirich V, et al. Prognostic role of CD8+ tumor-infiltrating lymphocytes
in stage III colorectal cancer
with and without microsatellite instability. Hum Pathol. 2004;35:808–816.
26. Naito Y, Saito K, Shiiba K, et al. CD8+ T cells infiltrated within cancer cell nests as a prognostic factor in human colorectal cancer
. Cancer Res. 1998;58:3491–3494.
27. Posselt R, Erlenbach-Wünsch K, Haas M, et al. Spatial distribution of FoxP3+ and CD8+ tumour infiltrating T cells reflects their functional activity. Oncotarget. 2016;7:60383–60394.
28. McCoy MJ, Hemmings C, Miller TJ, et al. Low stromal Foxp3+ regulatory T-cell density is associated with complete response to neoadjuvant chemoradiotherapy in rectal cancer. Br J Cancer. 2015;113:1677–1686.
29. Teng F, Mu D, Meng X, et al. Tumor infiltrating lymphocytes (TILs) before and after neoadjuvant chemoradiotherapy and its clinical utility for rectal cancer. Am J Cancer Res. 2015;5:2064–2074.
30. Lim SH, Chua W, Cheng C, et al. Effect of neoadjuvant chemoradiation on tumor-infiltrating/associated lymphocytes in locally advanced rectal cancers. Anticancer Res. 2014;34:6505–6513.
31. Yasuda K, Nirei T, Sunami E, Nagawa H, Kitayama J. Density of CD4(+) and CD8(+) T lymphocytes in biopsy samples can be a predictor of pathological response to chemoradiotherapy (CRT) for rectal cancer. Radiat Oncol. 2011;6:49.
32. Katz SC, Pillarisetty V, Bamboat ZM, et al. T cell infiltrate predicts long-term survival following resection of colorectal cancer
liver metastases. Ann Surg Oncol. 2009;16:2524–2530.
33. Nakagawa K, Tanaka K, Homma Y, et al. Low infiltration of peritumoral regulatory T cells predicts worse outcome following resection of colorectal liver metastases. Ann Surg Oncol. 2015;22:180–186.
34. Lee WS, Kang M, Baek JH, Lee JI, Ha SY. Clinical impact of tumor-infiltrating lymphocytes
for survival in curatively resected stage IV colon cancer with isolated liver or lung metastasis. Ann Surg Oncol. 2013;20:697–702.
35. Richards CH, Flegg KM, Roxburgh CS, et al. The relationships between cellular components of the peritumoural inflammatory response, clinicopathological characteristics and survival in patients with primary operable colorectal cancer
. Br J Cancer. 2012;106:2010–2015.
36. Coley WB. The treatment of malignant tumors by repeated inoculations of erysipelas: with a report of ten original cases–1893. Clin Orthop Relat Res. 1991;262:3–11.
37. Mlecnik B, Tosolini M, Kirilovsky A, et al. Histopathologic-based prognostic factors of colorectal cancers are associated with the state of the local immune reaction. J Clin Oncol. 2011;29:610–618.
38. Pagès F, Mlecnik B, Marliot F, et al. International validation of the consensus Immunoscore for the classification of colon cancer: a prognostic and accuracy study. Lancet. 2018;391:2128–2139.
39. Ma GF, Miao Q, Liu YM, et al. High FoxP3 expression in tumour cells predicts better survival in gastric cancer and its role in tumour microenvironment. Br J Cancer. 2014;110:1552–1560.
40. Martin ST, Heneghan HM, Winter DC. Systematic review and meta-analysis of outcomes following pathological complete response to neoadjuvant chemoradiotherapy for rectal cancer. Br J Surg. 2012;99:918–928.
41. Miyara M, Yoshioka Y, Kitoh A, et al. Functional delineation and differentiation dynamics of human CD4+ T cells expressing the FoxP3 transcription factor. Immunity. 2009;30:899–911.
42. Galetta H, Mansfield J, Richard B, Oguejiofor KK. Validation of multiplex immunofluorescence for use in analysis of tumour infiltrating lymphocytes. J Immunother Cancer. 2015;3:P411.
43. Martínez-Lostao L, Anel A, Pardo J. How do cytotoxic lymphocytes kill cancer cells? Clin Cancer Res. 2015;21:5047–5056.
44. Galon J, Pagès F, Marincola FM, et al. Cancer classification using the Immunoscore: a worldwide task force. J Transl Med. 2012;10:205.
45. Phillips SM, Banerjea A, Feakins R, Li SR, Bustin SA, Dorudi S. Tumour-infiltrating lymphocytes in colorectal cancer
with microsatellite instability are activated and cytotoxic. Br J Surg. 2004;91:469–475.
46. Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med. 2015;372:2509–2520.
47. Rozek LS, Schmit SL, Greenson JK, Tomsho LP, Rennert HS, Rennert G, Gruber SB. Tumor-infiltrating lymphocytes
, Crohn’s-like lymphoid reaction, and survival from colorectal cancer
. J Natl Cancer Inst. 2016;108:djw027.
48. Haikerwal SJ, Hagekyriakou J, MacManus M, Martin OA, Haynes NM. Building immunity to cancer with radiation therapy. Cancer Lett. 2015;368:198–208.
49. Dolan DE, Gupta S. PD-1 pathway inhibitors: changing the landscape of cancer immunotherapy. Cancer Control. 2014;21:231–237.
50. Sharabi AB, Lim M, DeWeese TL, Drake CG. Radiation and checkpoint blockade immunotherapy: radiosensitisation and potential mechanisms of synergy. Lancet Oncol. 2015;16:e498–e509.
51. Oh DY, Venook AP, Fong L. On the verge: immunotherapy for colorectal carcinoma. J Natl Compr Canc Netw. 2015;13:970–978.
52. Jacobs J, Smits E, Lardon F, Pauwels P, Deschoolmeester V. Immune checkpoint modulation in colorectal cancer
: what’s new and what to expect. J Immunol Res. 2015;2015:158038.
53. Xiang B, Snook AE, Magee MS, Waldman SA. Colorectal cancer
immunotherapy. Discov Med. 2013;15:301–308.
54. Gjertsen MK, Buanes T, Rosseland AR, et al. Intradermal ras peptide vaccination with granulocyte-macrophage colony-stimulating factor as adjuvant: clinical and immunological responses in patients with pancreatic adenocarcinoma. Int J Cancer. 2001;92:441–450.