Prognostic Tool for Immuno Age Debuts

Jérôme Galon, PhD

An assay that analyzes key elements of the tumor microenvironment in patients with colon cancer marks the first standardized method for evaluating an individual’s underlying immune system to be developed and sets the pace for tests in other malignancies that could be incorporated into conventional classification paradigms, according to a worldwide group of researchers who collaborated on the project.

The Immunoscore characterizes the number, density, and distribution of CD3-positive lymphocytes and CD8-positive cytotoxic T cells in the tumor core and invasive margins using a combination of automated immunohistochemistry testing and digital pathology. A patient can then be categorized as having a low, intermediate, or high Immunoscore depending upon preset parameters.

The assay has been validated in a study of tumor samples from more than 2600 patients with stages I-III colon cancer, according to results presented at the 2016 ASCO Annual Meeting in June.¹ The time to recurrence (TTR) was significantly longer in patients with a high Immunoscore, and the test was able to predict disease-free and overall survival. Additionally, a subgroup of patients with high-risk stage II colon cancer was identified through a low Immunoscore.

The Immunoscore breaks new ground in classifying cancers, said lead investigator Jérôme Galon, PhD, research director of the Laboratory of Integrative Cancer Immunology at the Inserm public research institute in France, who presented the results at ASCO. He also is co-founder of HalioDX, a diagnostic company seeking to commercialize Immunoscore.

“Today, there is not a single host immune characteristic that is taken into account for cancer patients. We don’t know anything about the immune system of a cancer patient because there is not a single standardized assay,” Galon said during the presentation. “In the era of immunotherapy, it is becoming essential to start classifying cancer patients based on immune parameters.”

As research on the new analytical tool moves forward, the Immunoscore could be used to enhance prognostic assessment and therapeutic management in a range of solid tumors, investigators have indicated.2 An assessment of a patient’s innate and adaptive immune responses could predict whether chemotherapy, radiotherapy, or checkpoint blockade immunotherapy agents would be effective.²

“This is a particularly timely finding in the era of immunotherapy, as Immunoscore-based assays could be used to predict which patients would be more likely to benefit from treatment modalities such as checkpoint blockade or whether strategies such as adjuvant therapy or cancer vaccines to prime immunity might be more appropriate,” said the Society for Immunotherapy of Cancer (SITC), which led the formation of the worldwide consortium that developed the Immunoscore.³

“More broadly, the results of the Immunoscore study have potential implications for the field of immune monitoring as a rapid means of determining response to treatment.”

The next step for the Immunoscore will be to incorporate the assay into randomized clinical trials “to stratify the patients based on what will be the first immune-based assay to measure the immune system of a cancer patient,” Galon said in an interview with OncologyLive.

Research is underway on Immunoscore tests for hepatocellular carcinoma and brain metastases. ² In July, HalioDx announced that the Immunoscore Colon test would be available to pathologists in Europe as a laboratory service and to researchers throughout the world by the end of this year.⁴

Bernard A. Fox, PhD, past president of SITC, said the Immunoscore is not yet ready to be incorporated into clinical practice but that assays evaluating the immune system to predict therapeutic outcomes probably would be introduced within the next 2 years. Fox is chief of the Laboratory of Molecular and Tumor Immunology at the Earle A. Chiles Research Institute at Providence Portland Medical Center in Oregon.

How Immunoscore Was Developed

“This is a great step but it’s still a first step and it’s a small step,” Fox said in an interview with OncologyLive. “There’s going to be additional information that we’re going to get in the next generation.”The Immunoscore findings presented at ASCO represent a milestone in an effort to develop a standardized assay that began more than a decade ago and required an unusual international partnership.

Galon noted that he and colleagues demonstrated the prognostic value of a patient’s preexisting immunity by quantifying immune cells, categorizing their location, and analyzing the impact on clinical outcomes in colorectal cancers in the mid-2000s.^5,6 Subsequent research built upon the concept of creating a method for evaluating immune system biomarkers.

Nevertheless, SITC faced considerable hurdles in forming partnerships that would help advance development of a classification system, Fox said. He said the organization approached nearly 20 companies, starting with those that had the advanced technology needed to conduct the analyses, about collaborating on the project.

“Nobody wanted to support it even though we were pointing out that if you knew you could stratify patients in your clinical trials, you may have drugs that worked but you took them off the shelf,” said Fox.

In 2012, SITC decided to help support the concept and began recruiting centers to participate. Ultimately, 23 pathology centers in 17 countries in North America, Europe, and Asia joined the study.

Evidence Presented at ASCO

“I’m very proud that the society pushed this,” said Fox. “I’m proud of the group of people who participated in this, for their generosity of their time and their treasure—all to benefit the patients.”In conducting the Immunoscore study, the participating centers collected tissue samples, performed staining on their slides, and then sent multiple consecutive slides and raw data to a reference center for testing and harmonization. The Mayo Clinic in Rochester, Minnesota, served as the external statistician.

For analysis, the Immunoscore employs computer technology and digital imaging, Galon explained. “There is software that is automatically counting every single immune cell that is infiltrating a tumor separately in the two tumor regions—the center and the invasive margin,” he said in an interview. “And then the software automatically calculates all the cells, all the cell density, and gives back the report of Immunoscore. It’s a fully automated process.”

Patients are stratified into 1 of 3 levels based on a score ranging from 10 to 14, depending upon the total number of high densities observed. 2 Both CD3 and CD8 are assessed in the tumor core and in the invasive margins.

The study criteria included patients with stages I/II/III colon cancer (T1-T4, N0-N2, M0) who had not received neoadjuvant treatment. In all, 3855 patients were evaluated for the Immunoscore but many did not meet the inclusion criteria for the study; the analysis was therefore conducted on the tissue of 2667 patients who were deemed eligible.

The participants were divided into a training set and two independent cohorts, an internal validation set and an external validation set. Galon said the arms were generally well balanced for age with a median of 68 years and tumor status, with 64.6% to 67.1% in each group at T3.

The proportion of patients with N0 nodal status was higher in the training and internal validation groups at 73.4% and 76.3%, respectively, than in the external validation group at 64.1%. The proportion of right-sided (proximal) and left-sided (distal) tumors was approximately even in the training and internal validation groups; right-sided tumors were more prevalent in the external validation cohort. The higher the Immunoscore, the better the prognosis. Twenty-six percent of participants had a high score, 49% had an intermediate score, and 25% had a low score. Patients were followed for recurrence for a median duration across centers of 5.9 years. Overall, participating centers counted more than 352 million CD3-positive T cells in the samples during the study, Galon said. The Immunoscore proved to be highly reproducible, with rates of whole-slide correlation between analysis by pathologist and assay at 0.98 for the center tumor region and 0.96 for the invasive margin region.

The study met is primary endpoint correlating TTR with the Immunoscore. The TTR was significantly longer in patients classified as Immunoscore-high versus Immunoscore-low in all 3 cohorts: the training subset of patients (HR, 0.41; 95% CI, 0.28-0.61), the internal validation set (HR, 0.41; 95% CI, 0.27-0.65), and the external validation set (HR, 0.51; 95% CI, 0.38-0.68). Those results were statistically significant for each group (P <.0001).

Galon said similar results were found for the secondary objectives of predicting disease-free and overall survival but did not provide details during his ASCO presentation.

Additionally, the Immunoscore lined up with the TTR for all 3 levels in a subset of patients with stage II colon cancer (n = 1433). The hazard ratio for this group was 0.36 with a 95% CI of 0.23-0.56, P <.0001.

A Note of Caution

“The Immunoscore assay, as we have demonstrated in this international study, has all the characteristics of a biomarker that can be done in routine practice,” Galon said. “It is pathology based, it is routinely feasible, it is reproducible, it is quantitative, it is standardized, and so, given the power of Immunoscore that we have demonstrated in this study, I believe it is now ready for clinical practice.”Amid the enthusiasm of the study team for the new assay, ASCO discussant Neil H. Segal, MD, PhD, wondered about the impact of microsatellite instability (MSI), which the Immunoscore does not specifically measure.

Microsatellites are short stretches of repetitive DNA that become unstable because of defects in the mismatch repair (MMR) system.

National Comprehensive Cancer Network guidelines recommend that MSI or MMR testing should be performed for all patients with metastatic colorectal cancer, patients with stage II disease because of the possibility that those with MSI-high scores would not benefit from adjuvant chemotherapy, and as part of Lynch syndrome screening.7 Immunohistochemistry is used for MMR, while polymerase chain reaction is used for MSI assessment.

Targeting Immunity

Segal, a medical oncologist at Memorial Sloan Kettering Cancer Center, said he does not think the Immunoscore is ready for clinical practice. “This is still a work in progress,” Segal said. “There is more information that we need in order to fully answer the question. The one caveat to interpreting this data is the contribution of MSIhigh colon cancers, which accounts for 15% of localized colon cancer.”The importance of the patient’s immune system in responding to therapy is at the heart of the SITC’s efforts to develop the Immunoscore. “The only thing that makes a difference in the life of a patient with metastatic cancer is their immune system,” said Fox. “There’s a lot of data in animal models that suggest that tumors that are not immunogenic—tumors that don’t have immune infiltrates—are not going to respond to checkpoint blockade, are not going to respond to costimulatory molecules.”

SITC researchers believe that the current reliance on the TNM staging system, with its tumor-focused methods of classifying all malignancies, has many shortcomings.3 Additional information can be gleaned from other aspects of tumor analysis such as cellular morphology and molecular pathways.

“However, instances in which clinical outcomes are drastically different between patients within the same stage, patients who maintain stable late-stage disease for years, or the rapid decline of early-stage patients, underscore the limited ability of the current staging systems,” the SITC said.³

During the past 5 years, the focus on the tumor cell itself in malignancies has evolved into an expanded understanding of the role of the tumor microenvironment, which Immunoscore researchers describe as “a set of cellular compartments comprising vascular, neuroendocrine, stromal, epithelial and immune cells.”² A growing body of evidence has demonstrated “a positive association between the density of intratumoral lymphocyte infiltrates in solid tumors and increased patient survival,” the investigators said.²

Specifically, the level of T cells expressing CD3 with CD8 or CD4 and memory T cells expressing CD45RO have been associated with outcomes in ovarian, head and neck, bladder, breast, liver, prostate, lung, melanoma, esophageal, and colorectal cancers.²

In designing the Immunoscore for colorectal cancer, researchers incorporated CD3 and CD8 testing because of prior research evidence about their significance but eliminated CD45RO because it is “highly overlapping” with T cell density and is difficult to include in the staining and slide process.² Fox believes much important information about the immune environment has been gained through the development of the Immunoscore that will be useful designing future therapies, particularly combinations that include immunotherapies. “What this points to is the fact that we need to have [agents] that are going to prime immunity,” he said.

Fox said the research also shows that pathologists employing emerging digital imaging technology are going to be a vital part of the cancer care team going forward. “The pathologists are going to be at the heart of oncology in the future,” Fox said. “They’re the ones who are going to be able to tell us what kinds of combinations to give patients.”

OncLive Web Editor Gina Columbus contributed to this report.

References

1. Galon J, Validation of the Immunoscore as a prognostic marker in stage I/II/III colon cancer: results of a worldwide consortium: final analysis on 3855 patients. Presented at: 2016 ASCO Annual Meeting; June 3-7, 2016; Chicago, IL. Abstract 3500.
2. KirilovskyA, Marliot F, El Sissy C, et al. Rational bases for the use of the Immunoscore in routine clinical settings as a prognostic and predictive biomarker in cancer patients. Int Immunol. 2016;28(8):373-382.
3. Society for Immunotherapy of Cancer. Society for Immunotherapy of Cancer Immunoscore Validation Project confirms prognostic value of immune-based classification system in patients with colon cancer. Targeted Therapies in Oncology. http://goo.gl/ yOyPys. Published July 26, 2016. Accessed August 24, 2016.
4. Immunoscore Colon test available [press release]. Marseille, Paris, France: HalioDX SAS. http://goo.gl/Y6A4j5. Published July 7, 2016. Accessed August 23, 2016.
5. Pagés F, Berger A, Camus M, et al. Effector memory T cells, early metastasis, and survival in colorectal cancer. N Engl J Med. 2005;353(25):2654-2666.
6. 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;131(5795):1960-1964.
7. National Comprehensive Cancer Network. Clinical practice guidelines in oncology: colon cancer. Version 2.2016. www.NCCN.org. Published February 18, 2016. Accessed August 23, 2016.