Flashcards in Neoplasia Deck (46):
Polycyclic aromatic hydrocarbons
Polycyclic aromatic hydrocarbons (PAH) are procarcinogens which require the action of hydroxylating enzymes such as aryl carbohydrate hydroxylase to become active carcinogens. These enzymes are ubiquitous, hence PAH are carcinogenic at their site of contact, but also absorbed into the blood stream and carcinogenic at distant sites e.g. kidney and bladder.
Explains why smoking tobacco is most strongly associated with carcinogenesis in tissues directly exposed such as the bronchus and larynx, smokers have a slightly increased risk of neoplasia in many other tissues.
Epidemiological studies have shown an increased risk of bladder neoplasms in workers in the rubber industry.
This has been found to be due to the aromatic amine, à-naphthylamine, which is converted into the active carcinogen 1-hydroxy-2-naphthylamine in the liver. Glucuronidation of this compound occurs in the liver, protecting the cells of the liver and other tissues from its carcinogenic effects. However, in the urinary tract, glucuronidase unconjugates the molecule, thus exposing the bladder urothelium to its carcino- genic effects.
The polycyclic aromatic hydrocarbons can act by adding alkyl groups to DNA, hence alkylating agents such as cyclophosphamide used as chemotherapeutic agents might also be carcinogenic.
While the risk is not sufficiently strong to contraindicate their use, there is evidence that patients treated with these compounds for conditions such as Hodgkin’s disease have an increased risk of developing a different type of neoplasm later in life.
These are an example of a class of compounds where recognition of their carcinogenic activity in laboratory studies has fortunately restricted their industrial use. For example, the dye dimethylaminoazobenzene causes liver cancer in rats.
Strongly carcinogenic in laboratory animals. It is not known to what extent they are carcinogenic in humans, but it is possible that generation of nitrosamines by fungi in poorly stored food could be responsible for some gastrointestinal neoplasms.
The association between asbestos and malignant mesothelioma (a neoplasm of the pleural, pericardial, or peritoneal mesothelial lining) is so strong that this disease is almost unknown in individuals who have not been exposed to asbestos.
Asbestos was a widely used building material because of its fire resistance, before the health risks of asbestos exposure were known. As a result of this, the incidence of mesothelioma continues to rise despite the restrictions now placed on the use of asbestos. There is also a strong link between asbestos exposure and carcinoma of the bronchus.
The mechanism responsible for the carcinogenic effect of asbestos is not known.
Exposure to nickel is associated with an increased risk of nasal and bronchogenic carcinoma. In the setting of haemochromatosis, iron could be said to be an indirect carcinogen in the liver; however, the development of cirrhosis is required before the increased risk of hepatocellular carcinoma in this con- dition can be realised.
In some parts of Asia, betel nut chewing substitutes for tobacco smoking as the preferred local vice. It has simi- lar hazards, as it is associated with an increased risk of the development of neoplasms of the oral cavity.
Helicobacter pylori infestation is a common cause of gastritis and peptic ulceration. Chronic Helicobacter pylori gastritis sometimes leads to intestinal metaplasia of the gastric mucosa. This results in the normal secretory epithelium of the gastric antrum being replaced by an epithelium with intestinal characteristics.
Sometimes this epithelium is well differentiated with a mixture of absorptive and goblet cells identical to those seen in the small intestine. In other cases the epithelium is less well differentiated, being identifiable as intestinal rather than gastric by the type of mucin that it produces. In the latter case, there is a small risk of the development of dysplasia and ultimately gastric carcinoma.
The association between Helicobacter pylori infestation and gastric carcinoma appears to be weak and presumably requires multiple cofactors. Nonetheless, this causative link has recently been confirmed in experimental animals infected with Helicobacter pylori.
There is a more direct link between Helicobacter infestation and a far less common neoplasm of the stomach, the mucosa-associated lymphoid tissue (MALT) lymphoma. It has been shown that despite having characteristics of a malignant neoplasm, such as clonality and invasiveness, MALT lymphomas sometimes regress when patients are treated with Helicobacter-eradicating antibiotics.
However, it is more likely that Helicobacter infestation represents a growth-sustaining stimulus, rather than a conventional carcinogen. These observations have led to some debate about whether MALT lymphomas are true neo- plasms or not.
Schistosomiasis is associated with an increased risk of carcinoma of the bladder. Interestingly, Schistosomiasis-associated bladder carcinomas are squamous carcinomas, rather than transitional cell carcinomas which are the usual type of malignant neoplasm of the bladder.
Clonorchis sinensis, the Chinese liver fluke, is also capable of inducing neoplasia of the bile ducts in which it dwells.
Some neoplasms such as carcinomas of the breast and prostate may require the presence of hormones to maintain or promote their growth.
There are also examples of abnormal exposure to some hormones being carcinogenic. For example, anabolic and androgenic steroids can cause the development of hepatocellular carcinoma, and oestrogens are associated with hepatocellular adenomas.
Certain rare tumours of the female genital tract, such as clear cell carcinoma of the vagina, are very strongly associated with in-utero exposure to diethylstilboestrol, which was used therapeutically during pregnancy in the past.
It is likely that there are many toxins produced by fungi that are carcinogenic. To date the best-established carcinogenic effect is that of the aflatoxins produced by Aspergillus flavus. These toxins occur as dietary contaminants and are linked to the high incidence of hepatocellular carcinoma in some parts of central Africa.
Malignant change in benign neoplasms
The majority of benign neoplasms do not alter in any way, but some benign neoplasms have the ability to progress to become malignant neoplasms. Probably the best-characterised example of this phenomenon is the adenoma-carcinoma sequence in the colon.
Adenomatous polyps of the colon are more numer- ous than colonic carcinomas, but all adenomatous polyps have the potential to develop into carcinomas, and many (but not all) carcinomas originate from adenomatous polyps. The polyps most likely to undergo malignant change show the greatest degree of histological dysplasia and a sequence of genetic changes that leads to the development of colorectal carcinoma from normal epithelium via adenomatous polyps has now been described
Neoplastic transformation of cells occurs in cells undergoing proliferation, and is particularly likely to occur if the cells are also undergoing metaplasia (defined and described in the previous chapter).
Neoplastic transformation of metaplastic epithelium usually follows a predictable and histologically identifiable sequence of low grade dysplasia progressing to high grade dysplasia/in-situ malignancy to invasive malignancy as additional genetic abnormalities are acquired in the neoplastic population. This progression is very well demonstrated in the cervix.
These are usually conditions characterised by high cell turn over a sustained period of time, usually resulting from a destructive form of chronic inflammation.
Congenital abnormalities can also be premalignant conditions: for example, maldescent of the testis is associated with an increased risk of testicular neoplasia in later life.
The carcinogenic process is the chain of events whereby a carcinogenic stimulus leads to the formation of a neoplasm. The principal steps in this process are as follows:
1. exposure of cell/tissue to carcinogen (initiation);
2. alterations to genes controlling cell growth and/or
3. irreversible change of growth control (persistence)
4. formation of neoplasm.
These four steps occur with decreasing frequency: exposure of cells to carcinogens is a very common event, and genetic alterations to growth-controlling genes probably occur quite frequently, but, because of inbuilt defense mechanisms, the latter two steps are relatively uncommon.
The division of the carcinogenic process into the stages of initiation, promotion and persistence is based upon experimental evidence from models of tumour formation in which initiating and promoting stimuli are required. However, our increasing understanding of the molecular genetics of this process indicates that the stages described above simply reflect the require- ments for more than one genetic change to occur before neoplasia becomes established.
Benign epithelial neoplasms are referred to as aden- omas if they consist of glandular (exocrine or endo- crine) cells, or papillomas if they have a papillary growth pattern – these are usually derived from a surface epithelium. Malignant epithelial neoplasms are referred to as carcinomas. This term usually has a pre- fix which refers to the pattern of growth or differenti- ation of the tumour, for example adenocarcinoma is the term used to describe a malignant epithelial neoplasm showing glandular differentiation. Often, a preced- ing adjective is used to describe the growth pattern or presumed cell of origin. In these situations the prefix ‘adeno’ may be dropped in common usage. Examples are papillary and follicular carcinomas of the thyroid (growth pattern) and ductal and lobular carcinomas of the breast (presumed cell of origin when these terms were coined, although now thought to be erroneous).
Benign mesenchymal neoplasms are named by combining a prefix describing their constituent cells with the suffix ‘oma’. For example, a lipoma is a benign neoplasm of fat, and an angioma is a benign neoplasm of blood vessels. In malignant mesenchymal tumours the suffix becomes sarcoma; thus a liposarcoma is a malignant neoplasm of fat, and an angiosarcoma is a malignant neoplasm of blood vessels (or, more strictly speaking, endothelium).
All neoplasms derived from lymphocytes are referred to as lymphomas, with the exception of those that circulate, which are referred to as leukaemias (e.g. chronic lymphocytic leukaemia, hairy cell leukaemia), and neoplasms of plasma cells, which are termed plasma- cytomas or myeloma depending on whether they affect single or multiple sites. The reason for the use of the blanket term ‘lymphoma’ is that the biology of these neoplasms is complex.
Lymphoreticular neoplasms: Lymphomas
Lymphomas are divided into Hodgkin’s disease and non-Hodgkin’s lymphomas. Hodgkin’s lymphomas are defined by the presence of the Reed-Sternberg cell, a morphologically character- istic cell of uncertain origin but probably derived from B lymphocytes.
Non-Hodgkin’s lymphomas exist in a bewildering diversity of forms which have spawned a number of different classifications. Broadly speaking, they can be subdivided into lymphomas of B lym- phocytes or T lymphocytes and high-grade or low-grade lesions, the latter distinction being the most important for management and prognosis.
More recently, certain types of lymphomas have become more strictly defined by cytogenetic or molecular genetic abnormalities. For example, mantle cell lymphoma is a type of B cell lym- phoma with morphology similar to low grade lymphomas but with a more aggressive clinical course. This type of lymphoma is characterized by a chromosomal translocation – t(11;14)(q13;q32) which leads to upregulation of cyclin D1, a cell cycle control gene.
For practical purposes all lymphomas are regarded as malignant, but some, such as nodular lymphocyte- predominant Hodgkin’s disease, have such a good prognosis that there is doubt as to whether they are true neoplasms.
Neoplasms of nervous tissue
Mature nerve cells very rarely give rise to any type of neoplasm; however, their precursors can give rise to a variety of tumours such as neuroblastoma and medul- loblastoma. These are examples of a variety of neoplasms bearing the suffix blastoma which are derived from embryonal cells and occur almost exclusively in children. Examples outside the nervous system include nephroblastoma (Wilm’s tumour) of the kidney and hepatoblastoma of the liver. Neuroblastomas occasionally show a remarkable property by maturing from a primitive, poorly differentiated tumour into a benign ganglioneuroma.
The vast majority of tumours occurring in the nervous system are derived from support tissues. In the central nervous system these are most commonly astrocytomas; in the peripheral nervous system they are derived from Schwann cells or nerve sheath fibrob- lasts, which form Schwannomas and neurofibromas, respectively.
Although they are usually sporadic and single, these benign nerve sheath tumours are notable for sometimes being multiple in the setting of the famil- ial syndromes of neurofibromatosis type 1 (multiple neurofibromas) and type 2 (acoustic Schwannomas, meningiomas and ependymomas).
An important concept that is illustrated by tumours of the central nervous system is the distinction between histological and biological malignancy. A non-invasive cerebral neoplasm acts as a space-occupying lesion and, therefore, has the potential to kill the patient, although it may do this over a longer period of time than its histologically malignant counterparts.
This term refers to neoplasms that either form from, or have characteristics of, cells of the amine and/or pre- cursor uptake and decarboxylation (APUD) diffuse endocrine system which consists of cells such as the islet cells of the pancreas, the calcitonin-secreting C cells of the thyroid, and the endocrine cells of the gut epithelium, or epithelial neoplasms that show evidence of this form of differentiation through the presence of neurosecretory granules within their cytoplasm.
These neoplasms can be benign or malignant, and are characterised by their ability to secrete peptide hormones or vasoactive amines. They usually present with symptoms caused by the substance that they secrete rather than symptoms directly attributable to the tumour itself. The resultant syndromes will be dis- cussed in more detail in the section below on clinical effects. The nomenclature of these neoplasms is vari- able and somewhat confused. However, it is common practice to refer to tumours secreting an identifiable product as causing a distinct syndrome according to their product, for example, insulinoma or gastrinoma; others are referred to by the generic term carcinoid. These tumours are generally of low to intermediate grade malignancy; their highly malignant counterpart is the so-called small cell carcinoma.
Benign proliferations of melanocytes are extremely common. These are known as melanocytic naevi. These are often congenital and may thus be hamartomas rather than true neoplasms (this distinction will be explained below), although others may be acquired in childhood or adulthood. Malignant melanocytic neoplasms are known as melanomas. Because this is a rather benign-sounding term, it is common practice to embellish this by referring to them as malignant melanomas (there is no such thing as a benign melanoma in humans, although they may occur in horses).
Germ cell neoplasms
Like other cell types, spermatogonia and oocytes are capable of forming neoplasms. The neoplasms that arise from them are generally diploid. In germ cell tumours arising in females, the sex chromosomes are invariably XX, whereas those arising in males can be XX or XY.
They may form neoplasms of essentially undifferentiated germ cells. In males these are referred to as seminomas; in females, dysgermino- mas. Tumours that show differentiation beyond this stage are known as teratomas (‘monster tumours’).
Mature teratomas occurring in females are common and benign, whereas immature teratomas are uncommon and malignant.
Germ cell neoplasms II
In males, teratomas of any type can give rise to metastases, although the more immature types are usually more aggressive. In mature cystic teratomas, well-formed squamous epithelium, glandular epithelium, neural tissue and teeth are frequently seen and almost any other tissue can be present.
The majority of germ cell tumours arise in the gonads, but some arise in sites such as the mediastinum and retroperitoneum, reflecting the site of origin and path of migration of the primordial germ cells.
Germ cell neoplasms III
The majority of teratomas in females are mature; in males the majority are immature. Malignant germ cell tumours are far more sensitive to radiotherapy and chemotherapy than, for example, malignant epithelial neoplasms. This has resulted in an excellent prognosis for seminomas and a relatively good prognosis for tera- tomas, even when metastatic disease is present.
A related group of neoplasms are the gestational trophoblastic tumours which are derived, as their name indicates, from placental trophoblast follow- ing a pregnancy. They are very uncommon following normal pregnancies, but are relatively more common following (hydatidiform) molar pregnancies. Like normal trophoblast, the cells of these tumours are well equipped to invade and metastasise, but are fortunately highly sensitive to chemotherapy.
A number of neoplasms show more than one neoplastic component, most commonly both epithelial and mesenchymal, indicating origin from a cell capable of differentiating down both lineages.
This is distinct from the recruitment of non-neoplastic stroma that occurs in most epithelial neoplasms (see section on tumour dependency). Examples of benign mixed neoplasms are the fibroadenoma of the breast and the so-called pleomorphic salivary gland adenoma.
Malignant neoplasms consisting of a mixture of epithelial and mesenchymal elements are generally referred to as carcinosarcomas; these occur most commonly in the female genital tract. There are some examples of mixed tumours which are distinctive clinicopathological entities such as synovial sarcoma (a misnomer because it is not derived from synovium) and the so-called pulmonary blastoma.
Poorly differentiated neoplasms
A proportion of malignant neoplasms do not show any evidence of differentiation, by conventional light microscopy. In the past these were assigned to the diagnostic dustbin of ‘anaplastic tumours’.
However, advances in electron microscopy and more particularly immunohistochemstry and cytogenetics now allow the majority of these neoplasms to be at least assigned to a broad category such as lymphoma or carcinoma, and sometimes to be diagnosed precisely.
These distinctions can be of great importance to patient man- agement: for example, an undifferentiated tumour that on further investigation proves to be a lymphoma may be highly responsive to appropriate chemotherapy.
Hamartomas are benign tumour-like lesions the growth of which is coordinated with that of the individual. They usually consist of one or more mature, well- differentiated tissue or cell types. Examples of such lesions are congenital melanocytic naevi (‘moles’) and pulmonary hamartomas. It should be noted, however, that there is not a strictly defined distinction between hamartomas and other benign neoplasms.
Choristomas are tumour-like lesions which consist of a perfectly formed mature tissue in an ectopic site. These are sometimes referred to as ‘rests’. Examples are ectopic adrenal tissue in the ovary, and ectopic pancreas in the wall of the gut. Like hamartomas, these are benign, non-neoplastic developmental abnormal- ities, the growth of which is coordinated with that of the individual in which they arise.
As is the case in all areas of medicine, we delight in applauding our fellows, and inevitably, many tumours have gained eponymous names. Most eponymously- named tumours also have a histogenetic label: for example, the Grawitz tumour of the kidney is more commonly known as renal cell carcinoma. However, some tumours, usually of obscure histogenesis, are known only by their eponymous name. The best- known examples are Ewing’s sarcoma of bone and Burkitt’s lymphoma.
There are several possible routes by which neo- plasms can metastasise:
• via lymphatics
• via the blood stream
• transcoelomic spread (across cavities such as the
• via the cerebrospinal fluid
• ‘seeding’ during surgery.
The process of metastasis can be broken down into the following stages:
• invasion of vessel/body cavity;
• homing to the ‘recipient’ organ or tissue; and
• establishment and growth of metastasis within the
In many senses, neoplasms have a parasitic relation- ship with the individuals in which they arise. In order to grow they need to be supplied with the nutritional requirements of any tissue, and although their growth is autonomous, for most neoplasms this term is relative and they retain some requirement for growth factor/ endocrine support. This relationship with the host has more than a purely academic significance; in some tumours it proves to be their therapeutic Achilles heel.
No solid tumour of normal cellularity can grow beyond a couple of millimetres in diameter without recruiting its own blood supply. The ability to do this is one of the few characteristics that most neoplasms have in common. However, precisely how they do so is not clearly established and may vary from neoplasm to neoplasm.
It is likely that most neoplasms can produce cytokines or growth factors that stimulate the proliferation and differentiation of endothelial cells, although there is also evidence to suggest that these factors may be produced by non-neoplastic accessory cells that are present within the tumour, most notably macrophages.
It is likely that tissue hypoxia is an important stimulus for angiogenesis. Apart from being biologically important in tumour development, inhibiting angiogenesis may also be sig- nificant in future treatment strategies for malignant neoplasms. Indeed, inhibition of angiogenesis may contribute to the efficacy of some existing chemotherapeutic regimes.
In common with the breast ductal epithelium from which they are derived, the growth of many breast cancers is also commonly influenced by oestrogens. The most effective medical treatments for breast cancer are agents that block the action of oestrogen on its intracellular receptors, such as tamoxifen, or that reduce oestrogen production, such as anastrozole.
The responsiveness of an individual breast cancer to anti-oestrogenic treatment can be predicted by the level of oestrogen receptor expression by the cells of the tumour.
Overall, oestrogen receptor-positive car- cinomas tend to be of lower grade and have a better prognosis than their oestrogen receptor-negative counterparts.
Prostatic epithelium is dependent upon androgenic stimulation for its growth and survival; castration of male rats results in apoptosis of prostate epithelial cells and partial involution of the gland. This characteristic is retained by most prostatic carcinomas, allow- ing their growth to be inhibited by androgen blockade. This can be achieved by orchidectomy (removal of the main source of androgens) or by interfering with the hypothalamo-pituitary-testicular axis.
Luteinising hormone (LH) released from the pituitary stimulates androgen production by Leydig cells in the testis. LH production can be inhibited either by treatment with oestrogens or luteinising hormone releasing hormone (LHRH) agonists which, after a transient stimulation, cause a long term depression of LH release.
Unfortunately, after a period of time, prostatic car- cinomas usually lose their androgen sensitivity, so the treatments described above are not curative.
Papillary and follicular carcinomas of the thyroid respond in the same way as normal thyroid follicular epithelium to the growth-promoting effects of thy- roid stimulating hormone (TSH). TSH suppression is used in combination with surgery and radio-iodine therapy to treat these neoplasms. It can be achieved by the negative feedback effect of replacement doses of thyroxine.
Local effects of neoplasia
The commonest clinical manifestations of the primary neoplasm are:
• a mass (palpable or visible);
• bleeding (due to ulceration);
• symptoms of irritation of the tissue of origin (e.g.
cough due to carcinoma of the bronchus);
• obstruction of a hollow viscus; and
• compression of or damage to adjacent structures,
Mass effects of neoplasia
This applies particularly to superficial lesions of tissues such as the skin and breast, although neoplasms aris- ing in deeper structures such as the stomach and colon may be palpable on clinical examination. Increasingly the presence of a mass manifests itself radiologically, rather than as a presenting complaint or on clinical examination, as modern radiological techniques such as CT and MRI scanning have high sensitivities for the detection of abnormal masses.
Paraneoplastic effects are those which occur in the presence of a neoplasm but which are not directly
Table 5.13 Examples of paraneoplastic syndromes
caused by the tumour itself or its metastases. These can be divided into the following categories:
• humoral (i.e. mediated by a secreted tumour product);
• immunological (usually tumour-associated autoimmune phenomena); and
• uncertain cause
Paraneoplastic syndrome: Humoral
Syndromes caused by the secretion of an ‘appropriate’ product, such as insulin by insulinomas or catechol- amines by phaeochromocytomas, have been described above. Many other neoplasms secrete ‘inappropriate’ products such as ACTH or antidiuretic hormone by bronchogenic carcinomas. The commonest of these syndromes is humoral hypercalcaemia of malignancy, which is caused by secretion of parathyroid hormone- related peptide (PTHrP). The normal function of this peptide, which is produced by many epithelial tissues, is not known. It causes hypercalcaemia by binding to the parathyroid hormone receptor in kidney and bone. It is possible that, to some extent, the weight loss and cachexia associated with many malignant neoplasms is due to secretion of catabolic factors, possibly cytokines such as tumour necrosis factor-alpha and interleukin 6, but it is more likely that in most cases it is multifactorial.
Paraneoplastic syndrome: Immunological
Autoimmune disease can be triggered by malignant neoplasms. A significant proportion of patients with dermatomyositis, particularly those developing it for the first time later in life, have an underlying malig- nancy. Also, membranous glomerulonephritis, a cause of nephrotic syndrome, which results from immunological damage to the glomerular basement membrane, can be initiated by an underlying neoplasm. A variety of neurological syndromes can be associated with malignant neoplasms, most commonly small cell carcinoma of the bronchus. Examples of this are the myasthenia-gravis-like Eaton-Lambert syndrome, cere- bellar ataxia and dementia. The majority, if not all, of these syndromes seem to be caused by the produc- tion of autoantibodies against nerve cell components such as neurotransmitter receptors, possibly due to an immune response to antigens shared by tumour cells and neurons.
In order to be able to establish a screening programme, it is necessary to fulfill the fol- lowing criteria.
• There must be a diagnostic test available that can practically be applied to large numbers of people and can be repeated on different occasions in the same individual.
• The test must have high levels of sensitivity (the proportion of tests carried out in individuals
who have the disease that detect the disease) and specificity (the proportion of positive tests that are due to the disease, rather than other diseases or artifacts).
• The test must give comparable results between different testing centres.
• It must be possible to apply the test to a high proportion of the target population.
• There must be established and effective treatments for the disease.
• There must be evidence that screening for the disease in question can reduce levels of morbidity and mortality from that disease.
Cervical carcinoma screening
Squamous carcinoma of the cervix and its precur- sor lesion, cervical intraepithelial neoplasia (CIN), can be detected by exfoliative cytology. Given that these diseases affect a relatively accessible site in a relatively young age group, and that by the time cervical carcinomas become symptomatic, they are often locally advanced, screening for cervical carcinoma has been widely practised for many years. However, only recently has a co-ordinated national screening pro- gramme been developed in the UK.
The test employed, microscopic examination of exfoliative cytology samples stained by the Papanicolaou technique, is highly sensitive and specific. It has the considerable advantage of allowing detection of the precursor condition, CIN, and thus allowing simple local curative treatment to be carried out before invasive carcinoma develops. The major disadvantage is that it is labour-intensive and therefore expensive and prone to subjective error. However, despite this, the UK cervical carcinoma screening programme appears to be successful, because prevalence and death rates from cervical carcinoma in the UK are declining. It is probable that this is attributable to the screening programme, but this cannot be stated with absolute certainty as the screening programme represents (for obvious ethical reasons) an uncontrolled experiment.
Although cervical screening has been highly successful, even in the best circumstances, it is not completely effective. Also the applicability of the systems of cervical screening used in countries such as the UK to developing countries is questionable. Testing for infection by high risk HPV types is currently under investigation as an alternative method of screening.