Hair follicles can develope tumors too (trichoblastic carcinoma). All cells can develope cancer (some of them with a very low frequency); it is something inherent in our genome and biology.

I think the frequency is basically due by cell proliferation rate and mutagens exposure. But I am sure there are some differences (probably epigenetic programs) that influences the capacity to acquire the capacity of proliferate without control.

The only cells that can't form cancer AFAIK are the ones that can't ever proliferate. Erythrocytes are so differentiated that they loose their nucleus. Therefore, without any genes, cancer can't emerge from them. Their precursors however, which possess a nucleus, can develop cancer.

Some cells form cancer more often than others. Different parts of the body are exposed to different concentrations of various carcinogens. Tissues have different proliferative potential, some cells do not proliferate physiologically (cardiomyocytes), while others proliferate like crazy (epithelium). The more "breaks" are limiting cell's life cycle, the more barriers need to be overcome by mutations in protooncogenes, suppressors, and apoptotic genes to start uncontrolled cell division and eventually lead to cancer.

In case of the myocytes do not undergo cell division unless there is an injury and therefore very less cell division, the chances of mutation and cancer are less frequent.

Further,heart is also not directly exposed to carcinogens except for those in the circulating blood that leads to the decreased risk of mutations in the cardiac cells. Thus, because of less cell division of cardiac cells, heart is at a less risk of developing cancer.

Most cancers come from epithelial cells; due to increased rates of cell division. These type of cancers are called carcinomas.

As I know cancer, it is inherently the disease of the cell cycle. What is important to initiate tumor formation is a mutation either in a set of tumor suppressor genes or proto-oncogenes. Once the mutations have occurred, it is important that these be heritable.

If certain cells are terminally differentiated and are least likely to divide any further, mutations in those cells will not give rise to cancer. Therefore, cancer is most likely seen in cells that divide very fast in the body because they are not only prone to mutations but also to them being inherited in the progeny cells.

Hope this helps!!

Staying within the carcinomas ( epithelial malignancies ), these are very rare in some anatomical sites (and when present, they are in the context of some syndromic complex), epididymis , seminal vesicles , uterine tubes , the small intestine etc. Conversely locations where cancers arise most frequently are the transition areas where active chronic inflammatory processes result in a high cells turnover (sometimes with metaplastic processes ) . In these circumstances carcinogens ( chemical and biological ) have more effect ( the bronchial mucosa, the cervix where there is a transition between squamous and columnar epithelium , the gastric mucosa which changes and becomes absorbent type ( intestinal metaplasia ), etc. . Other areas of transition are particularly subject to turnover following the sequence of hormonal secretory and proliferative stimuli ( ductulus in the breast , at the junction between the duct and lobule where the majority of breast cancers occurs , the endometrium etc). Probably this is not the sole cause of the difference in the incidence of cancer in various epithelia, but it can be one of the main switchboards. Seems to me that, for example, the seminal vesicles have been studied to find some intrinsic characteristic to justify their inability to develop cancer , but I think with few results.

In other circumstances specific genes are likely involved. The keratoacanthoma of the skin that arises from structures of the hair. It shares many morphological and molecular characters with squamous cell carcinoma of the epidermis, however, are probably still active genes that regulate hair growth (anagen and catagen). Thus, this tumor grows and regresses on schedule by the hair cycle.

Neurons and RBC wont switch in cancer cell as neurons can't able to replicate and differentiate. Whereas RBC are anucleated which avoid the chances of DNA damage and mutations.

From a theoretical standpoint, I believe that all cells have the potential to become "cancerous" due to mutation events.

Even cells that do not have proliferating capabilities, changes in transcriptional machinery may allow these cells to undergo cell division and potentially become cancerous.

As is has been stated in other answers, there needs to be a perturbation (mutations, cell damage, inflammation) in order to promote a de-differentiation of a cell into a cancerous one. Taking into account Waddington's epigenetic landscape, the magnitude of the perturbation needed should be directly related to how differentiated the cell is in its lineage.

Most often actively replicating precursor cells readily produce cancers (blood cancers are common). Rarely dividing cells produce cancer less often (nerve cell origin).

I guess bascially every cell could potentially turn into a cancer cell once enough gene aberrations / mutations arise that lead to a full transformation (note: cancer is a multi-hit process, for which multiple events are needed). Having said that, the chance to turn into a cancer cell is than bigger for those cell that rapidly proliferate (as the genetic changes more rapidly precipitate lead to genomic instability) or cells that physiologically undergo genetic breaks (recombination events in lymphocytes).

Perhaps ery's (but not their precursors) are an exception. One has also to realize that what we finally see as cancer cells is the result of selection processes. A popular idea is that cancer is a kind of micro-evolution process, implying that selective forces / advantage impact on the final result, i.e. whether the cancer appears or not. In that sense, what we see as cancers may not completely reflect the presence of cells on their way to become cancer cells.

Hope these thoughts contribute.

reading all those comments where you only explain cancer by mutations, which is derived from the somatic-mutation-theory (smt) that says that the cells normal state is quiescence and mutations are the origin of cancer, i feel that there is a missing part.

There are also other views like the tissue-organization-field-theory (TOFT) that states that the natural state of the cell is proliferation and motility and that this is only inhibited by its microenvironment (stroma, ecm, cell-cell communication, a lot of nice works in this direction come from mina bissels and Ana Sotos labs). If you answer the question in context to this theory it might be thinkable that every cell that in theory is able to divide and communicates with its environment might be able to initiate malignant growth when under the right conditions. This would probably exclude RBCs because they dont proliferate anymore due to a missing nucleus.

Just wanted to add my 2 cents for a more "wholistic" view.

In theory, any cell type can give rise to cancer. Some adult cell types are non-proliferating (for example neurons), but their less differentiated precursors can become cancerous. Having said that, it is intriguing that some cell types rarely give rise to cancer whereas others frequently do so. A good example is the glandular cells of the 2 largest organs of the male reproductive system. The prostate gland is the site of the most frequent gender-specific cancer in males. In contrast, the seminal vesicles, which at least superficially do not seem that different and are located immediately adjacent to the prostate, hardly never develop cancer. It is not because the seminal vesicles harbor some magic substance that inhibits cancer growth because prostate cancers readily infiltrates into seminal vesicles. For some unknown reason, the epithelial lining of seminal vesicle glands, unlike that lining the prostatic glands, does not become cancerous.

Currently it well known that cancer affects only dividing cells, that means adult differentiated cells are far from developing cancer. Because cancer is the disease of the genetic material (DNA), it must be expected that all nucleated cells should be involved to have cancer even the adult cells. Differentiated cells may change to become blast cell if some mutations occur to affect genes (protooncogenes) responsible for cell division. Therefore all nucleated cells may develop cancer but the dividing cells are more susceptible for that.

The proliferating capacity and mutagens play role of course. But mutagens can't affect cells that are not proliferating and are sciencent. So I would say that the primary or essential requirement for carcinogenesis is the DNA synthesis and cell division.

Most cancers seem to behave consistently with the Cancer Stem Cell hypothesis. Since most tissues are a dynamic flow of stem cells that proliferate only to maintain their numbers and differentiate to produce the functioning cells of the organ or tissue, cancer might most readily come from mutations (or epigenetic changes, that express growth not called for by the rest of the tissue) in these cells. However, the Yamanaka experiment showing conversion of a fibroblast into an iPSC shows that under the right conditions any nucleated call can become a stem cell and possibly a cancer stem cell. The reason so few tumor cells, when first taken from the patient, grow in vitro may be due to only a tiny fraction of them being cancer stem cells and even fewer are in the supportive niche/microenvironment that allows them to grow. Established cell lines are totally different, and probably not good cancer models.

Frequent DNA replications and the ensuing cell divisions lead to acquiring and accumulating mutations. This lead to activation of proto-oncogenes and loss of tumor suppressor genes, which are the primary reasons for cell transformation.

This is one of the advantages of having most cell divisions done by amplifying transit cells, the progeny of which will (mainly) be discarded after their useful lifespan. E.g., the haematopoietic stem cell does one division and a daughter is pushed out of the niche. It then becomes susceptible to erythropoietin and becomes an erythroblast. The erythroblast undergoes 7 divisions, producing 128 erythrocytes (red blood cells). Thus the stem cell protects us from a much higher rate of cancer that would occur if most of the replications were done by the stem cells.

Certainly. Stem cells undergoing symmetric cell division do good for an organism. The transit amplifying cells formed during asymmetric cell division specify the lineage, multiply rapidly and are prone for transformation.

There is some cells which are not able to proliferate. These kind of cells are in the highest level of differentiation and are not able to reprogram back to proliferating cell line.

So these kine of cells can't cause cancer

Aref,

Not quite. If the cell has a nucleus, it has the same genome as it's less differentiated precursors and hence can be reprogramed. The Yamanaka gene set in a transfecting plasmid can make almost any cell an iPSC.

The question is quite interesting. It has to be viewed from multiple dimensions to answer this question. The concept of accumulating mutations in differentiated cells or cancer stem cell theory (discussed above by others) may answer to an extent and concludes that cell needs nucleus and it should divide. However, to add complexity, epigenetic changes in cells also lead to cancer. These epigenetic changes may be influenced by surrounding microenvironment. Further, formation of teratomas (an unusual form of tumor) when embryonic cells were injected into mice does not need carcinogens or mutations and may be influenced by microenvironment. The complex mechanisms are still unclear.

Ravindra is correct. Epigenetic changes can also lead to cancer, so gene sequencing will not provide all the answers. Gene expression (RNA) arrays also are problematic because most of the tumor's cells are not the Cancer Stem Cells (CSCs) of interest (see below). The microenvironment or niche is crucial for CSC proliferative potential. In 1965 we showed that in planarians, rapid fissiparic proliferation led to teratomas, presumably due to inappropriate normal tissue signaling when the wrong tissues came into contact. These teratomas were not malignant and animals that had divided away from the tumor had the same probability of getting a new one as the normal population.

In our Hybrid Spheroid Assay, using tumor cells taken directly from human patients, we obtained a plating efficiency (tumor stem cell fraction) of 0.5% for cervical cancer CSCs. In monolayer culture, not a single cell grew when 40,000 of these tumor cells were plated, yet 200 colonies would have been expected if a niche were not required. Our Hybrid Spheroids provide this niche (or at least a niche-like environment sufficient to allow CSCs to survive and grow). (see Radiation Research Society annual meeting and International Congress for Radiation Research abstracts. Several papers have been submitted and/or are in preparation on this topic.

Any cell may become a cancer cell. The question is what is the likelyhood that it will happen + what factors promote the change to a cancerous phenotype.

Regarding original question: While cancers of the prostate are common, cancers of the accessory Cowpers gland or bulbourethral gland are very rare. This is worth investigating, since both are presumably androgen dependent??

Cancer can come from any dividing cells. Differentiated cells are inert and possess specific functions but they are not the source of cancer unless they are mutated to activate oncogenes or inactivate suppressor gene. Erythrocytes cannot be the source for cancer because they have no nucleus.

Any cell can become cancerous. The problem is when and where...and WHY???

Look at it from epidemiological perspective (all numbers are rough but true from the top of my head):

no direct environmental exposure (excude in utero): pediatric cancers - 20% brain tumors, 30% sarcomas and 45% leukemias; 5% others. Frequency in kids population about 15:10.000 (0.015%). Extremely rare epithelial cancers. Genetics? sure,but in about 10%.

Adults: full environmental exposure, accruing with time (age). frequency in population: reaching 75%.

Dominating types: epithelial cancers, 99%: gut, lung, breast , prostate form 95% of all malignancies, which will make 99% if skin cancer is added. Genetics? less than 1%. So why all energy/money go into genetics? If someone can give me a sound answer please.

...rarest cancer ever? interesting question, definitely deserves an exhausting literature search.

Nikita gave wide answer and needs more explanation. If any cell can gives cancer, why then we need hard effort to establish cancer cell line in vitro or in vivo. Why do we try to use strong carcinogenes to induce cancer in animal? not all cells have the ability to bring cancer. Cells with dividing ability may have this ability.

Under normal conditions tumour suppressor proteins such as p53 and p16 regulate DNA duplication, the cell cycle and apoptosis through negative feedback mechanisms. If damage to DNA during mitosis is not repaired, such genetic instability will induce cell death via activation of these tumour suppressor proteins and their downstream paths. Mutations to, or deletions of, the genes that transcribe these tumour suppressor proteins lead to the loss of the protective constraint on cell division and regulated cell and results in immortality. In itself, this is not a problem. However, over time further genetic mutations may now be allowed to occur. Unrepaired mutations in key signalling genes constitutes one of the early steps in the progression to a malignant phenotype. Such mutated, cancer-promoting genes (oncogenes) may give rise to mutated signalling proteins that are constitutively active and may thus not require any external signals to promote growth. Even more dangerous is the inability to react to the usual negative feedback mechanisms, allowing for a dedifferentiated state, constant pressure to proliferate (cancer formation) or constant pressure to migrate (invasiveness).

The microenvironment of each cell type (other cell types, the extracellular matrix and secreted components) also plays a major role in the suppression of cancer progression. Over time these restricting factors may also be altered or diminished, reducing their influence over cancer progression.

I hope this sheds some light.

Cells from mole mice never produce cancer. Actually this is a interesting animal model for cancer study, especially when focusing on tumor microenvironment. It seems that hyaluronan plays a major role in  suppressing  cancer progression in mole.

Has anyone compared rate of DNA repair efficiency with incidence of cancer?  There should be an inverse correlation. I wonder if epithelial cells have poor efficiencies? Perhaps DNA repair rate is only 1 of many factors..still, it  would be interesting to see a clear correlative study on this. 

Cancer is considered systemic disease. And every cell has genetic program that determine the life course of such cell.and crells in the body under cretin circumstances. Genetic error. Environmental effect. Switch on and switch off gene may develop cancer.

Re.: Venil Sumantran's answer, I would suggest that repair fidelity us more important than repair rate. High rate with low fidelity should be very cancer prone, while low rate may result in higher lethality and hence lower carcinogenicity. Also, one should consider rate times time available for repair (set by cell's metabolism) as the parameter to estimate amount or fraction of damage repaired.

To C. Lange: How does one measure fidelity? Would ds breaks be repaired with less fidelity than ss breaks? Perhaps there are special proteins which monitor fidelity. Please do say more about this important point. Thanks.

Wonderful discussion and a very nice question,Just adding some points to the discussion. With the development in high throughput techniques and specific molecular, biological, clinical research coming together, many aspects of cancer initiation-formation, biology, progression are unravelling such as cell type specificity, stem cell origin of cancer, transcriptions factors for pluripotency vs differentiation, genetic errors and mutations, epigentic and immunological factors along with socioeconomic factors, ethnicity, and geographical location (diet) regarding the incidence and understanding of cancer.

I think the cells that are most prone to cancers depends on their exposure to the hits involved in cancers and also their proliferative and differentiation status.

1. So epithelial cancers are the commonest such breast, prostate, lung, colorectal.

2. then Melanoma, Lymphoma, Leukemia, the list varies with the location in the globe, ofcourse.

3. Neuoblastomas is a rare cancer that occurs in the children, while most brain tumours develop from cells that support the nerve cells.

4. And many rare tumours that are found in the 'rare list' are very specific such as Merkel cell skin cancer, Angiosarcoma of the heart. Very well differentiated cell types are may be not switched back to pluripotency easily or may be these cell types not easily exposed to triggering mechanisms as are the cancer stem cells.

I would like to bear in mind the basic cell types and the location of their stem cell type in our body keeping in mind their normal functionality and derailment from it, by DNA damage and also their normal proliferative and apoptotic signals (as the damaged cells are may not be cleared) regulates how common or how rare a tumour may be. Ofcourse other factors mentioned above do come into play.

How a stem cell becomes a cancer stem cell having the multipotent capacity by acquiring genetic modifications is vital, as clonaity of cancer has shown that origin of the tumour goes back to a primary cancer stem cell. It is important to be hit by one or multiple mutations by triggering agents in the environment (bacterial, viral or toxic) and the cells proliferative capacity allowing it not to be able to repair the damage or other enzymes or molecules in the repair pathways being ineffective. Also, epigenatic mechanisms may come into play along with the genetic aberrations in the mutihit process of cancer formation and progression.

The refs below in cancer biology, progression and pluripotency factors may help further understanding.

1. http://www.nature.com/nature/journal/v432/n7015/full/nature03094.html

2. http://www.sciencedirect.com/science/article/pii/S1044579X12000570

2.Cell. 2006 Aug 25;126(4):663-76. Epub 2006 Aug 10.

Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors.

Takahashi K1, Yamanaka S.

Dr. Keckesova discovered the lactb tumour suppressor mechanism asking the same question as you did. And here's the result:

https://www.nature.com/nature/journal/v543/n7647/abs/nature21408.html

Cancer can be produced only in dividing cells. Differentiated cells that are non dividable cells cannot change to cancerous cells. Moreover, I believe that cancer occurs only in stem cells that can be found in every tissue even brain and heart. These stem cells proliferate continuously to substitute dead cells in every tissue, therefore when any genetic errors occurred in these cells, that lead to loss of division controlling,may change the cell to cancer cell.     

Every living and dividing cells had potential to produced cancer. Cancer not occurred in simple way. Cells should experience various mutation that lead into cell proliferation, disruption of DNA repaired, escaping from immunity system, and many more. After experience a lot of complex mutation, therefore the cells will have their cancer phenotype. Even though, the mutation it self is not enough, their environment should be suitable so they can lived.