CDX2 -- New Prognostic Biomarker in Stage II Colorectal Cancer

Alexander M. Castellino, PhD

January 20, 2016

A new biomarker, the CDX2 protein, can identify a small subgroup of patients with stage II colorectal cancer (CRC) who are likely to benefit from adjuvant chemotherapy, say researchers. At present, these patients are usually treated with surgery alone.

Use of CDX2 as a prognostic biomarker for these patients may be more predictive than clinical staging, which typically guides treatment decisions in these patients, says an expert not involved in the research.

The findings come from a study published online January 21 in the New England Journal of Medicine.

Researchers from Stanford University report how they identified CDX2 ― a master transcriptional factor involved in intestinal development ― and found that it could be used as prognostic biomarker in CRC.

Patients with CDX2-negative CRC had lower 5-year disease-free survival (DFS) compared with those with CDX2-positive tumors. In addition, the analysis showed that the 5-year DFS was higher for patients with CDX2-negative stage II disease who received adjuvant chemotherapy compared with those who did not receive adjuvant chemotherapy.

"CDX2 has identified a group of patients with stage II CRC who will benefit from adjuvant chemotherapy," principal investigator and corresponding author Michael F. Clarke, MD, of Stanford University's Institute for Stem Cell Biology and Regenerative Medicine, told Medscape Medical News.

"Dr Clarke and his group have to be saluted for this work," C. Richard Boland, MD, from the Baylor University Medical Center, Dallas, Texas, told Medscape Medical News. He coauthored an editorial that accompanied the study.

A second accompanying editorial highlights the methology employed, which involved data sharing. Dan L. Longo, MD, and Jeffrey Drazen, MD, write: "The new investigators arrived on the scene with their own ideas and worked symbiotically, rather than parasitically, with the investigators holding the data, moving the field forward in a way that neither group could have done on its own."

For their study, Dr Clarke and colleagues used the National Center for Biotechnology Information Gene Expression Omnibus (NCBI-GEO) database to identify the CDX2 gene. Tissue samples were provided by the Cancer Diagnosis Program of the National Cancer Institute (NCI-CDP), the National Surgical Adjuvant Breast and Bowel Project (NSABP), and the Stanford Tissue Microarray Database (STMD).

"The findings have generated a new hypothesis that is now ready for testing in a prospective randomized clinical trial," Dr Longo and Dr Drazen add.

Cancer Stem Cells in Colorectal Cancer

For Dr Clarke, this study culminates many years of collaborative work and "is the first to demonstrate the probable therapeutic benefit of cancer stem cell research."

It began with laboratory interest in markers to define and enrich colon cancer stem cells. Coauthors Piero Dalerba, MD, now at Columbia University, and Debashis Sahoo, PhD, now at the University of California, San Diego, made significant contributions to this work, which first started with identifying the activated leukocyte cell adhesion molecule ALCAM/CD166 as a marker in immature colon epithelial cells expressed at the bottom of colon crypts.

ALCAM/CD166 was capable of identifying colon cancer stem cells, which led the researchers to look for genes that were not expressed in ALCAM/CD166-positive tumors but that were expressed in ALCAM/CD166-negative tumors. The candidate genes would be inversely associated with the stem cell–like state and more highly with differentiated cells.

"We reasoned that those proteins would likely be involved in the maturation of colon tissue and might not be found in more aggressive, immature cancers," Dr Sahoo said in a Stanford University press release.

To do this, they used bioinformatics and interrogated more than 2300 human colon gene expression arrays, downloadable from the NCBI GEO repository, and identified 16 candidate genes. However, they were interested only in genes that were actionable, that is, genes that were translated into proteins that could be visualized histochemically in clinical specimens.

The proverbial needle in the haystack was one gene, CDX2, for which there already was a clinical-grade diagnostic test.

"We chose CDX2 because it was the only candidate that was already used as a diagnostic biomarker in the clinic," Dr Dalerba said in a Stanford University press release. "However, we were also intrigued by the fact that CDX2 is a master transcription factor controlling the expression of many differentiation genes in colon epithelial cells," he added.

From the Lab to the Clinic

To determine the clinical relevance of CDX2, the Stanford researchers evaluated CDX2 association with 5-year DFS in a discovery set of 466 patients from the NCBI-GEO database, which had information both on 5-year DFS and CDX2 status.

They found significant differences between the patients with CDX2-negative tumors (n = 32) and those with CDX2-positive tumors (n = 434): the 5-year DFS was 41% vs 74% (P < .001).

For the validation set, the researchers analyzed microarrays of colon cancer tissue from the NCI-CDP.

Microarrays were selected to reflect a balance of patients with and without tumor recurrence and patients with a long-term follow-up.

CDX2 expression was determined from immunohistochemistry with a CDX2 monoclonal antibody that was previously validated for diagnostic use. Again, patients whose tumors were negative for the CDX2 protein (n = 38) had a worse prognosis. Five-year DFS was 48% vs 71% for patients with CDX2-positive (n = 318) tumors (P < .001). Overall survival was also lower — 33% vs 59% (P < .001).

In both the discovery and validation datasets, the association remained significant regardless of age, sex, tumor stage, and tumor grade.

Patients with stage II disease and CDX2-negative tumors also had a significantly lower 5-year DFS in both the discovery (49% vs 87%; P = 0.003) and validation sets (51% vs 81%; P = .004).

Adjuvant Chemotherapy Benefit

The researchers next examined whether CDX2 status could be used to identify patients with stage II disease who might benefit from adjuvant chemotherapy.

For this analysis, the researchers pooled databases of historic cohorts of patients with stage II and stage III disease (n = 1897). Patients were pooled from the NCBI-GEO (n = 222) and NCI-CDP (n = 265) from the discovery and validation sets, respectively, with two expansion cohorts — 1216 patients from the NSABP C-07 trial, and 194 patients from the STMD.

Of 1897 patients, 669 had stage II disease. Most of them were CDX2-positive (n = 621), of whom 389 received adjuvant chemotherapy. The remaining 48 patients were CDX2-negative, and 23 of these received adjuvant chemotherapy.

Adjuvant chemotherapy was associated with a higher rate of 5-year DFS in both stage II and stage III patients:

  • Stage II: 91% vs 56% for no chemotherapy (P = 0.006)

  • Stage III: 74% vs 37% for no chemotherapy (P = 0.001)

Most significant was the observation that DFS for patients with CDX2-negative tumors was superior when they received chemotherapy compared with those who did not receive chemotherapy.

The benefit of adjuvant chemotherapy was seen regardless of depth of invasion of the primary tumor, number of lymph nodes resected at surgery, and the number of metastatic lymph nodes.

"Our results indicate that patients with stage II or stage III CDX2-negative colon cancer might benefit from adjuvant chemotherapy and that adjuvant chemotherapy might be a treatment option for patients with stage II CDX2-negative disease, who are commonly treated with surgery alone," the researchers write.

Adjuvant chemotherapy might be a treatment option for patients with stage II CDX2-negative disease, who are commonly treated with surgery alone. Dr Michael Clarke and colleagues

"Prior studies to identify a subgroup of patients with high-risk stage II colon cancer have not been robust, and the lack of prognostic and predictive criteria underscores the need to discover biomarkers than can facilitate the selection of patients for additional treatment," Dr Boland and coauthor Ajay Goel, PhD, write in their editorial.

"The prognostic implication of CDX2 is an important and significant finding of the study, with implications for clinical practice," Dr Boland told Medscape Medical News. "While all patients with stage III disease are candidates for adjuvant chemotherapy according to the guidelines, there is an issue of the standard of care for patients with stage II disease," he added.

"The benefit of adjuvant chemotherapy in patients with stage II has been debated, since recurrence never develops in most of these patients," Dr Boland told Medscape Medical News.

Dr Boland explained that in the current treatment landscape, 5-year DFS is marginal for patients with stage II disease who receive adjuvant chemotherapy. Depending on the registry or trial, the 5-year DFS for stage II patients is about 80% to 85%, and although some patients appear to benefit from adjuvant chemotherapy, the increment of benefit is less than 5% overall. But everyone who receives adjuvant chemotherapy is at risk for complications (eg, neuropathy associated with oxaliplatin) for only a modest improvement in outcomes, he indicated.

"The current standard for clinical prognostication relies principally on pathological staging," Dr Boland and Dr Goel write in their editorial. Although the current study is not perfect or definitive, it "provides an opportunity for oncologists to move beyond what has been an inadequate method of selecting patients with stage II colon cancer for adjuvant chemotherapy."

"This retrospective study requires prospective confirmation with uniform interventions, which was not necessarily the case in the different cohorts, Dr Boland and Dr Goel write in their editorial.

"[These] findings raise the important question of what the mechanism might be at work in silencing CDX2; the answer to this question could lead to the discovery of new approaches to treating fundamental problems,” they add.

The Stanford researchers agree. "Given the exploratory and retrospective design of our study, these results will need to be further validated," they write.

Nonetheless, Dr Clarke told Medscape Medical News that the similar DFS rates obtained with the discovery and validation sets across multiple datasets provide grounds for confidence that that the effect seen is likely to be significant notwithstanding the small number of patients who had CDX2-negative disease.

Dr Boland acknowledged that prospective studies take years to complete, and he predicted that clinicians will not wait that long. "With the availability of CDX2 as a diagnostic test, clinicians are likely to use CDX2 to identify stage II patients who may benefit from adjuvant chemotherapy," he told Medscape Medical News.

The authors and editorialists have disclosed no relevant financial relationships.

N Engl J Med. Published online January 20, 2016. Full text, Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: