A comprehensive genomic analysis of rectal cancer in a large cohort of patients who undergone neoadjuvant treatment and had available detailed clinicopathological annotations highlighted clinically relevant associations between transcriptomic features, immunologic profiles, and treatment outcomes. APC mutations were less frequent in the lower than in the middle and upper rectum, which could explain the more aggressive behaviour of distal tumours. No somatic alterations had significant associations with response to neoadjuvant treatment in a treatment-agnostic manner, but KRAS mutations were associated with faster relapse in patients treated with neoadjuvant chemoradiation followed by consolidative chemotherapy. Overexpression of IGF2 and L1CAM was associated with decreased response to neoadjuvant treatment. RNA-sequencing estimates of immune infiltration identified a subset of microsatellite stable (MSS) immune hot tumours with increased response and prolonged disease-free survival (DFS). Findings from the integrative analysis of genomic and transcriptomic profiles are published by Walid K. Chatila, PhD candidate at Memorial Sloan Kettering Cancer Center in New York, NY, US and colleagues on 15 August 2022 in the Nature Medicine.
This study included 738 patients from four cohorts: 25 patients from the ACOSOG Z6041 study, 71 patients from the TIMING study, 163 patients treated at Memorial Sloan Kettering (MSK) who underwent sequencing using research protocols, and 479 patients who underwent sequencing as part of their care at MSK. Patients were profiled using a combination of DNA and RNA sequencing platforms.
Tumours located in the rectum comprise one-third of colorectal cancers. Although the overall incidence of colorectal cancer has decreased in the past decades, the incidence of rectal adenocarcinoma is increasing in patients younger than 50 years old. Despite being historically grouped as a uniform disease, colon and rectal cancers are managed as distinct clinical entities and have different recurrence patterns and associated morbidity.
Patients with locally advanced rectal cancer are treated with neoadjuvant treatment, consisting of radiation and chemotherapy, to reduce the risk of local recurrence and increase the likelihood of sphincter preservation. Most rectal cancers show some degree of response to neoadjuvant treatment, ranging from minimal response to complete tumour eradication. A complete response is associated with low recurrence rates and excellent survival, challenging the benefit of total mesorectal excision surgery and creating opportunities for organ preservation in up to 50% of patients with locally advanced rectal cancer.
The authors wrote in the background that the wide variability in reported rates of complete response cannot be fully explained by macroscopic tumour characteristics, such as tumour location, size and stage, nor by differences in treatment alone. The identification of biomarkers of response to neoadjuvant treatment may facilitate prognostication and the development of strategies to increase the proportion of responders.
Genomic analyses of colorectal cancer have disproportionately focused on colon tumours, due in part to the difficulty in obtaining untreated rectal samples. Furthermore, these tumours cannot be used to analyze genomic differences across rectal segments because their precise location was not annotated. This is in contrast to the distinction between left-sided and right-sided colon cancer. By focusing on anatomically verified and clinically annotated rectal tumours, the present data provide a unique opportunity to investigate the role of genomics in the tumourigenesis of rectal cancer.
While prior studies have linked molecular features such as KRAS mutations to treatment resistance, the genomic determinants of response to neoadjuvant treatment have not been systematically studied. Previous analyses have been limited by small sample size. Large-scale sequencing efforts focused on untreated surgical resection samples due to the scarcity of treatment-naive specimens. In this study, the investigators analyzed DNA and RNA extracted from preoperative endoscopic biopsies from tumours located all along the rectum with comprehensive clinicopathological annotations to investigate the biology of rectal cancer.
In the article published in the Nature Medicine, the authors report an integrative analysis of genomic and transcriptomic profiles in 738 patients diagnosed with rectal adenocarcinoma. They first delineated the molecular landscape of somatic alterations across different tumour stages, highlighting biologically relevant differences by rectal location that could better illuminate previously observed variations in prognosis. They then identified somatic alterations, transcriptomic features and immune infiltration profiles that correlate with differences in clinical outcomes for patients with locally advanced rectal cancer treated with neoadjuvant treatment.
The results show that APC mutations are less frequent in tumours from the lower rectum. This provides a genomic basis for clinical differences that had traditionally been explained based on anatomical considerations alone.
Prediction of response to neoadjuvant treatment for locally advanced rectal cancer is clinically relevant because it may help select patients for organ preservation. The group had previously reported an association between KRAS mutations and lower rates of complete response, based on data that came predominantly from the TIMING study. By contrast, that association was not observed in the larger cohort of patients described in the latest article.
In terms of recurrence, KRAS mutations were associated with shorter DFS in patients treated with neoadjuvant chemoradiation followed by consolidative chemotherapy. However, underlying differences in patient accrual, treatment guidelines and choice of surgical versus non-operative management could act as confounding factors. Furthermore, elements such as the time interval between completion of neoadjuvant treatment and surgery may also affect tumour response and patterns of recurrence. In order to robustly characterise the predictive and prognostic role of KRAS mutations, larger cohorts need to be analyzed, ideally within the context of prospective, randomised clinical studies.
The role of the tumour microenvironment, and specifically the immune infiltrate, in locally advanced rectal cancer response to neoadjuvant treatment is highlighted by dMMR/MSI tumours, which are characterised by high lymphocyte infiltrate and respond well to immune checkpoint inhibitors (ICIs). However, recent clinical studies have failed to show a benefit of ICIs in the general locally advanced rectal cancer population, which consists of approximately 95% of pMMR/MSS tumours.
In this work, the investigators have identified pMMR/MSS immune hot tumours with extensive immune infiltration and favourable outcomes. This is consistent with previous observations, such as those that led to the development of the Immunoscore assay as a measure of pan-cytotoxic lymphocyte densities in the tumour-associated stroma that can be used as a prognostic factor in non-metastatic colorectal cancer.
The immune hot pMMR/MSS tumours in the present cohort have higher levels of cytotoxic lymphocytes and T cells, as well as expression of immune checkpoint-related proteins, than the dMMR/MSI set. The fact that these immune hot tumours accounted for only approximately 8% of the pMMR/MSS cases may explain why immune checkpoint blockade has overall not shown activity in pMMR/MSS patients. The distinct immune profile described could be used to select pMMR/MSS patients for inclusion in future studies with ICIs.
Future research may also involve novel strategies to induce immune infiltration in IG1 and IG2 tumours. The present data show that the overexpression of IGF2 and L1CAM correlates with poor outcomes in locally advanced rectal cancer. If validated in prospective cohorts, IGF2 inhibitors could be explored as a therapeutic strategy to elicit response in these patients. The present results emphasize the clinical relevance of L1CAM as a marker of poor response to neoadjuvant treatment in locally advanced rectal cancer.
The authors commented that their work has some limitations. The majority of samples were sequenced with targeted-exome panels; the use of broader sequencing platforms might yield valuable additional insights. RNA was available only for a small subset of patients.
The results from this analysis of associations between genomic features and response to neoadjuvant treatment need to be interpreted with caution and will need to be further investigated in future cohorts. The authors anticipate that the molecular profiling of specimens collected through recently completed clinical studies such as OPRA, TNT and PROSPECT will be valuable to further explore this question.
Reference
Chatila WK, Kim JK, Walch H, et al. Genomic and transcriptomic determinants of response to neoadjuvant therapy in rectal cancer. Nature Medicine 2022;28:1646-1655.