Molecular Biology in the Pathogenesis of Musculoskeletal Neoplasia
Molecular Biology in the Pathogenesis of Musculoskeletal Neoplasia
Michael J. Monument, MD, MSc, FRCSC, FAAOS
Kevin B. Jones, MD, FAAOS
Dr. Jones or an immediate family member serves as a board member, owner, officer, or committee member of Alex Lemonade Crazy 8 and the Clear Cell Sarcoma Foundation. Neither Dr. Monument nor any immediate family member has received anything of value from or has stock or stock options held in a commercial company or institution related directly or indirectly to the subject of this chapter.
ABSTRACT
Neoplasia is considered to result from dysregulation of cells at the molecular level. However, the aberrant molecular mechanisms that initiate and promote progression to a malignant state have been identified across the full breadth of molecular biology. It is important to review molecular pathology at levels of the DNA and genome, transcription to RNA, translation to protein, posttranslational chemical modifications of proteins, protein stability in cells, signaling cascades and pathways, tumor cell intrinsic evasion of the immune system, and the tumor microenvironment. Examples of musculoskeletal neoplasia resulting from derangement of each of these layers of molecular biology should be considered.
The central dogma of molecular biology was that the DNA of genes was transcribed into RNA, which was translated into amino acid sequences, which folded into proteins. It was long thought that this process worked only in that direction, until study of HIV and other retroviruses proved the capacity of certain natural enzymes to generate DNA from RNA. Nonetheless, these layers provide a reasonable organization of any consideration of the molecular biology of neoplasia. Varied classes of regulatory molecules that govern the transitions between these states are also worthy of consideration as applied to the field of cancer biology.
MAKING A CANCER DIAGNOSIS
When a mass is referred to as a cancer, it is grouped with entities that have previously demonstrated a capacity to spread to distant sites in a dangerous manner. Categorizing a particular tumor in a given patient under a certain diagnosis does not mean the tumor will or has metastasized. The heterogeneity in behavior observed among all the masses assigned any specific cancer diagnosis suggests the existence of heterogeneity on the molecular level that, once identified, may offer more specific diagnoses, treatments, and prognostic information. This is the hope of the movement termed personalized medicine.
Much has been learned in recent years about the heterogeneity of the cells that comprise a single malignancy.1 Furthermore, heterogeneity among the human hosts of any cancer type will also affect each malignancy’s behavior and biologic potential. The molecular biology of cancer therefore works to bring into focus both the molecular underpinnings of oncogenesis and the heterogeneity in the malignancies it produces.
Cancer cells use the same tools as normal cells to regulate their biology. Some fundamental differences in the products of that regulation (or goals, although intent seems a tough-to-resolve concept for a cell) have been named the hallmarks of cancer.2,3 The hallmarks are, essentially, the characteristics by which cancer cells are distinguished from their noncancerous counterparts.
The first hallmark is self-sufficiency in growth. Normal cells depend on signals from their environment, including neighboring cells, the extracellular matrix, and factors diffusing from local sources or from the vasculature after delivery from distant, systemic sources. Cancer cells achieve independence from external growth signals by aberrant expression of both the ligand and the receptor of some growth signaling pathway, by expression of a mutated growth receptor that activates the downstream pathway without the need for a signal from the ligand binding, or by overexpression—or normal expression of an overactive variant—of one of the downstream factors of that growth pathway.
Normal cells stop growing either by senescence (permanent exit from the cell cycle) or apoptosis (programmed cell death). Cancer cells evade internal and external signals that would otherwise prompt senescence or apoptosis. It is impossible to distinguish these capacities entirely from the self-sufficiency of growth because many of the involved pathways overlap. These pathways include evasion of contact inhibition, loss of organizing polarity, and silencing of both immune-mediated and intrinsic apoptosis in response to genetic or chromosomal instability.
Even if induced to grow independently from external signals and internal signals prompting senescence or apoptosis, most cells can only copy or replicate their DNA (a critical portion of the cell cycle) so many times before the telomeres, or caps on the ends of chromosomes, have been shortened to a length that will no longer permit stability and continued growth. Some very aggressive malignancies manage to silence apoptotic responses to the instability derived from critically shortened telomeres, but most other malignancies must activate some mechanism of lengthening the telomeres to continue growing in a limitless manner.
Another hallmark is that the metabolism of cancer cells can differ markedly from that of other cells. For example, many cancer cells use inefficient glycolysis alone, rather than the full Krebs cycle and the mitochondrial electron transport chain to generate energy in the form of adenosine triphosphate. Nonetheless, even metabolically abnormal cancer cells require a constant supply of nutrients and removal of waste products. Because many cancer cells inefficiently burn through glucose much faster than other cells, they are even more dependent on an excellent blood supply than other cells. To address this need, the cells of most solid tumors induce and encourage angiogenesis, the ingrowth of blood vessels.
Invasion is another hallmark of cancer cells. The invasion of basement membranes or other tissue planes defines many malignancies on a histopathologic basis. Before a cell can spread to a distant site and begin growing, it must free itself from the physical constraints of its initial environment and enter the bloodstream.
The final capacity to metastasize formally depends on additional abilities to survive in the bloodstream, cross another basement membrane after landing somewhere else, and then grow in the new and unfamiliar environment. This last requirement may be the most stringent and most rarely attained.
The original article on the hallmarks of cancer article was published in 2000, but an update in 2011 emphasized a new hallmark, immune evasion.2,3 Of course, consideration of musculoskeletal neoplasia more generally than just sarcoma must include entities that acquire only one or a few of these hallmarks.
THE MOLECULAR BIOLOGY OF MALIGNANCIES
The 2011 report on the hallmarks of cancer also emphasized certain molecular tools by which malignancies achieve their phenotypic hallmarks. Specifically noted were genomic instability, inflammation, reprogramming of metabolism, and interactions with the microenvironment (Figure 1).
FIGURE 1 Illustration showing the hallmarks of cancer.
Extensively studied and validated, these quintessential biologic programs are maladapted by altered mesenchymal lineage cells in the process of malignant transformation and manifestation of sarcomatous disease. Understanding how these various processes promote sarcomagenesis aids in disease characterization, detection, and hopefully the introduction of new treatment modalities.
Each cancer cell’s capacities result from two principal inputs: the cell of origin and the transformation process that changes that cell into a malignant cell. Any cellular phenotype or differentiation characteristic discernible in a cancer cell was once presumed to reflect the cell of origin alone, and the transformation to achieve the hallmarks of cancer was considered an entirely separate or even dedifferentiating series of events. Any malignant cell producing bone matrix was assumed to derive originally from an osteoblast; any cell producing cartilaginous matrix, from a chondrocyte; any cell staining for smooth muscle markers, from a smooth muscle myocyte; and any cell expressing keratins, from an epithelial cell.
Researchers have recently become aware that transformation may powerfully affect the differentiation phenotype of the resultant malignancy. For example, synovial sarcomas are not thought to arise from epithelial cells, and yet most of them express cytokeratins.
The inverse is also true, and may be especially critical to the understanding of transformation in mesenchyme. Innate capacities of certain cells of origin may contribute to the hallmarks themselves. Because most human malignancies are—and an even greater proportion of most human cancer biologists study—carcinomas, or cancers of epithelial tissues, generalized philosophical considerations such as these hallmarks may not fully apply to cancers derived from mesenchyme. Many normal mesenchymal cells innately possess the capacity to move through basement membranes or even survive in the bloodstream or other tissues. A popular theory in the study of epithelial cancers is that they must become more like mesenchyme during the process of gaining metastatic potential.4 These innate mesenchymal characteristics do not necessarily eliminate the difficulties of full-fledged cell transformation, but may make the process easier.
It is important to review the molecular biology related to cells undergoing malignant transformation. The central dogma serves as a foundation for the discussion, and examples from musculoskeletal neoplasia are considered.
CHANGES AT THE DNA LEVEL
At the level of the genetic code in some malignancies, some genes are simply missing or deleted. These changes may arise from the loss of an entire chromosome or the start site or a critical coding section of a gene. Large sections of chromosomal loss or gain have been detectable for decades using formal cytogenetics. Newer technologies using hybridization, or the adherence of complementary DNA sequences to each other on a microscale (single-nucleotide polymorphism arrays) or macroscale (fluorescent in situ hybridization), or direct sequencing of partial or entire genomes have revealed much more than was previously known about important smaller losses and gains in the copy number of genes.
The loss of some genes grants new capacities to some malignancies; such genes are called tumor suppressors because their native suppression must be lost for a tumor to develop (Figure 2). A cell must lose both copies of most tumor suppressors to gain any related cancer hallmark, but the loss of even one copy of other genes will lead to haploinsufficiency (phenotype driven by the insufficient expression—but not complete loss—of a gene from loss of one of the two copies of the gene in the human genome), which can enable new cell behaviors (Table 1).
FIGURE 2 Illustration showing the oncogenes and tumor suppressor genes in sarcoma.
Genetic alterations can result in activated genes that promote uncontrolled cell cycle progression and cellular proliferation (oncogenes) or impair and/or silence genes normally responsible for cellular growth arrest, genome stability, and cellular death programs of damaged cells (tumor suppressor genes). In most instances, expression of oncogenes and inhibition of tumor suppressor gene function are not mutually exclusive and coevolve together via selective cellular pressures and progressive genomic instability throughout the course of sarcoma progression.
Musculoskeletal neoplasia, although it comprises less than 1% of all malignancies, played a major role in the identification of the two most important tumor suppressor genes, RB1 and TP53.5 Osteosarcoma, the most common bone sarcoma, and undifferentiated pleomorphic sarcoma (previously called malignant fibrous histiocytoma), the most common soft-tissue sarcoma, are both characterized by TP53 and RB1 dysfunction.6 Osteosarcoma is also the second most common malignancy in patients with hereditary retinoblastoma. Although these patients inherit one mutated copy of RB1, a critical portion of their remaining functional copy is often lost or deleted in the process of osteosarcomagenesis. Both TP53 and RB1 are cell cycle checkpoint regulators, which means that their loss can enable growth that is independent from external signals and free the cells from the senescent or apoptotic signals that would normally result from DNA damage.
After these two genes, the next most frequent genetic locus lost in malignancies is CDKN2A, which contains two overlapping genes, INK4A and ARF. Interestingly, these also disrupt the RB1 and TP53 checkpoints, respectively. CDKN2A is the most common locus lost in the peripheral chondrosarcomas that arise secondary to osteochondromas.7ATRX is another gene, involved in genome integrity by regulating repair of DNA damage that is often silenced in sarcomas.8
PTEN is another tumor suppressor gene commonly deleted in sarcomas.8 Unlike TP53, RB1, and CDKN2A, loss of only one allele of PTEN can contribute to oncogenesis by shifting apoptosis pathways and enhancing growth signals.9 There are many genetics means of manipulating the PI3′-lipid signaling pathway that is regulated by PTEN protein.9
Additional copies of other genes contribute to transformation in other malignancies. These oncogenes require no mutations or alterations to make them dangerous, only additional gene copies. Two groups of musculoskeletal neoplasms are characterized by amplification of almost the same region of chromosome 12, usually appearing as an extra ring or rod chromosome.10 Although recently sequenced to identify all elements contained therein, three genes—CDK4, TSPAN31 (formerly SAS), and MDM2—have long been recognized to have extra copies on these extra chromosomes. Excess CDK4 protein silences RB1; excess MDM2 silences TP53. The two tumor types are atypical lipomatous tumors (also called well-differentiated liposarcoma or lipomalike liposarcoma) and parosteal osteosarcoma.
Although copy number losses and gains at certain genes or chromosomal regions characterize different musculoskeletal neoplasms (Table 2), tumors can be generally grouped into genomically stable neoplasms, which have few copy number variations, and genomically complex and unstable neoplasms, which have rampant copy number variations and significant heterogeneity among cells within each tumor. Osteosarcoma is the typical prototype of genetically complex and unstable neoplasms.11