FormalPara Key Points

  • Indigenous peoples of the Americas can develop liver cancer at an early age in the absence of cirrhosis.

  • Integrative genomic analysis has uncovered peculiarities both in gene expression and epigenetic reprogramming, revealing a divergent molecular subtype of liver cancer.

  • This divergent molecular subtype of liver cancer is associated with infection by an autochthonous sub-genotype of the hepatitis B virus.

The idea that ancestry modulates the molecular determinants of cancer is relatively new in genomic research and requires further study [1]. While the scientific question is fascinating, the health issue is alarming since Indigenous peoples remain starkly under-represented in cancer genomics studies, effectively excluding them from the benefits of such research [2]. Our work has shed light on this phenomenon in cancer patients with Indigenous American backgrounds. Specifically, we have identified a strong correlation between Indigenous American ancestry and the development of an early-age subtype of hepatocellular carcinoma (HCC), the main form of primary liver cancer [3].

HCC is one of the most prevalent and deadly tumor types worldwide. It usually afflicts individuals in middle and old age after protracted liver diseases, such as cirrhosis. Considerable efforts have been dedicated to elaborating a clinically relevant molecular classification of HCC [4]. However, a unifying classification that includes the whole heterogeneity of HCC will remain an ongoing concern—if not an unattainable goal—as long as some populations, such as Indigenous peoples, are under-investigated [2]. For instance, the early-age onset of non-cirrhotic HCC presented in a significant fraction of Indigenous American patients originating from Alaska and the Andes who are infected with the same autochthonous sub-genotype F1b of the hepatitis B virus (HBV) represents an illustration of this concern [3, 5].

Accordingly, our main objective has been to characterize the early-age form of non-cirrhotic HCC developed by Indigenous peoples of the Americas at the genomic level based on a cohort of Native Andean patients from Peru. It should be noted here that Peru is, with Bolivia, one of the countries in the Americas with the highest level of Indigenous genetic structure in its general population [6]. Around 80% of Peruvians self-identify as either Andean-Amazonian Indigenous or mixed-Indigenous ancestry.

Methods

An integrative analysis of gene expression and DNA methylation in non-cirrhotic HCC developed by Native Andean patients, including adolescents and young adults, was conducted to understand the molecular correlates and determinants of the disease in this population. Native Andean patients were recruited while receiving treatment for liver tumors at the National Cancer Institute of Peru (INEN) according to Institutional Research Ethics Committee Protocol 10-05.

Biobanking

One of the primary challenges we faced was accessing biological samples from cancer patients with Indigenous backgrounds. Accordingly, to promote the inclusion of individuals from under-represented Indigenous peoples in genomic studies, we developed biobanking activities, which led to the creation of the National Tumor Biobank of Peru in 2022 (Peruvian Law 31336). Hosted at INEN, this facility is firmly involved in collecting biological samples as well as sociodemographic and clinical data from cancer patients of the Andean-Amazonian Indigenous communities of Peru and their descendants, in accordance with local laws governing Indigenous and personal ownership of data (Peruvian Laws 27811 and 29733, respectively).

Indigenous Ancestry, Cancer Genomics, and Virus Detection

The Indigenous American ancestry of the cancer patients was determined by their mitochondrial DNA haplotype. Human mitochondrial DNA haplotypes are used by population geneticists to trace the matrilineal inheritance of modern humans and their distribution around the globe. Most of the members of the Indigenous American communities typically carry one of the four ancestral lineages labeled haplogroups A, B, C, and D [7]. Through integrative genomics, we then evaluated gene expression, pathway analyses, and DNA methylation in the early-age HCC developed by Native Andean patients [3]. In parallel, we conducted an in-depth molecular analysis of HBV genomes and viral loads using ultra-sensitive molecular assays, as described previously [8].

Results and Discussion

According to transcriptome analysis, HCC is normally divided into two classes, i.e., proliferative and non-proliferative, based on differences in pathway activation, phenotype, and prognosis (Fig. 76.1) [4]. However, we reported that Indigenous American haplogrouping is associated with peculiarities in gene expression in Native Andean patients who develop non-cirrhotic HCC around 20 years of age [3]. While Andean HCC falls roughly into the proliferative class, it also exhibits idiosyncratic traits in additional signaling pathway activation (Fig. 76.1). From a molecular standpoint, Andean HCC features a divergent subtype between the proliferative and non-proliferative classes, usually regarded as mutually exclusive according to the molecular classification of HCC (Fig. 76.1). Additionally, Andean HCC displays high levels of DNA methylation, contrasting with the global hypomethylation pattern considered a hallmark of HCC [3]. In this regard, our findings uncover an original biological model of epigenetic reprogramming in cancer development.

Fig. 76.1
A Venn diagram. The circle titled Native Andean intersects proliferative and non proliferative class. The intersection of proliferative class and Native Andean reads H B V and higher A F P level. The intersection of Native Andean and non proliferative class reads H C V and lower A F P level.

A schematic representation of HCC molecular classification, integrating Andean liver tumors. According to prognostic gene signatures and signaling pathway activation, Andean specimens fall into a nontypical position within the classification

Full size image

As HBV is suspected to be the prominent risk factor for HCC in South America, we performed a comprehensive molecular study of HBV infection in Native Andean patients. Intriguingly, HBV infection in Native Andeans is associated with a very low viral DNA burden, disclosing a significant rate of occult infections [8]. This observation sharply contrasts with the prevailing paradigm that relates higher HBV DNA loads to the onset of HCC, considered “early,” at around 40 years of age [9]. A phylogenetic analysis of the HBV genome clustered every isolate within the sub-genotype F1b, a viral clade infecting historically Indigenous American peoples [10]. Such high prevalence rates of occult infection with autochthonous HBV clades (F, G, and H) have been documented to a large extent in the Indigenous peoples of the Americas [5, 11]. It is noteworthy that Native Alaskans from the Yupik tribe with early-onset HCC associated with the same sub-genotype F1b have also been described [5]. Our assumption is that the HBV-associated carcinogenic process might differ substantially in Indigenous populations of the Americas from that generally observed in other populations. In this view, the distinctive tumor dynamics could result from an incomplete adaptation of Indigenous Americans to the autochthonous HBV sub-genotypes [12].

Our findings reveal a major role for bio-anthropology in molecular oncology, with the characterization of a clinically relevant molecular subtype of HCC in patients of Indigenous American descent. Overall, our research represents the first integrative genomic characterization of a molecular subtype of cancer that preferentially affects people of Indigenous ancestry. Our study stresses the necessity of conducting further bio-anthropological research programs with Indigenous people to meet their specific needs in cancer detection, prevention, and treatment. In this regard, we will continue to collaborate with medical anthropologists to improve our understanding of the socio-economic and cultural factors influencing access to care for Andean-Amazonian people.

We are grateful to all patients whose participation was essential to this research. This work was supported by ITMO Cancer of the French National Alliance for Life Science and Health (Aviesan) and the French National Cancer Institute (INCa) with funds administered by the French National Institute of Health and Medical Research (Inserm), grant agreement 21CD025-00. This work was also supported by ANRS | Emerging Infectious Diseases on funds administered by Inserm, grant agreement AO 2021-1 CSS12.