Lack of differentiation or anaplasia is considered the hallmark of malignant transformation. Anaplasia meaning reversion from a higher level of differentiation to a lower level of differentiation and it is marked by pleomorphism, abnormal nuclear morphology with increase Nuclear/Cytoplasmic ratio, loss of polarity of cells. Pleomorphism i.e. variation in size and shape that implies: some cells may be many times larger than their neighbors and others may be extremely small and primitive appearance. So as such there is no direct relationship between cell size and malignant potential as malignancy is mutifactorial and it varies among different types of cancers.

As I am working on lung cancer cells, which are different types like small-cell lung cancers (SCLC), non-small-cell lung cancers (NSCLC) and other types. The three main subtypes of NSCLC are adenocarcinoma (most common), squamous-cell lung carcinoma, and large-cell lung carcinoma. The NSCLC are slow to metastasize but have poor response to therapy as compared to small-cell lung cancers (SCLC) which are highly malignant with higher metastasis rate but they show high initial response to Chemotherapy.

Regards.

The above query is completely contextual and basic. Shedding of vestigial matter! from the cells to focus on division seems to be a hyper speculation! and does not address the malignancy to size correlation. Size to Malignancy are two extreme parameters; specifically Malignancy is a multifactorial dependent outcome which may or may not increase or decrease the cell size. This cannot be generalized to size of cells as every cell in a tumor is different.

There's a very interesting paper that came out recently that addresses a part of this question. To my knowledge this is the only paper I came across that addresses this question both in vitro and in vivo in a rat model. The main finding is that large cells, which actually stain positive for beta-gal (senescence) and are multinucleated are more malignant than their 'normal' counterparts. This work was done in a murine fibrosarcoma cell line and both types of cells (large vs. normal) were implanted separately into a rat model. The tumors that developed from a single large cell were more malignant than those that developed from a single normal cell.

I hope this helps.

Attached is the paper.

Below is the paper.

Article Formation of Solid Tumors by a Single Multinucleated Cancer Cell

I believe that you cannot generalise a link between cancer cell size and malignancy over all cancer types, as there are highly malignant cancer subtypes that have small cells, but also some large cell variants of specific cancers that are very aggressive.

Lack of differentiation or anaplasia is considered the hallmark of malignant transformation. Anaplasia meaning reversion from a higher level of differentiation to a lower level of differentiation and it is marked by pleomorphism, abnormal nuclear morphology with increase Nuclear/Cytoplasmic ratio, loss of polarity of cells. Pleomorphism i.e. variation in size and shape that implies: some cells may be many times larger than their neighbors and others may be extremely small and primitive appearance. So as such there is no direct relationship between cell size and malignant potential as malignancy is mutifactorial and it varies among different types of cancers.

As I am working on lung cancer cells, which are different types like small-cell lung cancers (SCLC), non-small-cell lung cancers (NSCLC) and other types. The three main subtypes of NSCLC are adenocarcinoma (most common), squamous-cell lung carcinoma, and large-cell lung carcinoma. The NSCLC are slow to metastasize but have poor response to therapy as compared to small-cell lung cancers (SCLC) which are highly malignant with higher metastasis rate but they show high initial response to Chemotherapy.

Regards.

U must compare survival rates of acute melanoma and slow growing prostate cancer

In our works what does really matter is the form, the shape of chromatin and not the cell size. The Fractal Dimension (FD) is rather more important for malingnancy than other parameters as chords, perimeter or size.

Karel Smetana

In general there is no correlation between malignancy and cell size. It also depends on the type of tumour. However, rapidly dividing cells are blastogenic, and in general blasts are larger than their normal (resting) cell counterpart.

At least for the hematological cell lines I've not seen any soever such correlation. I think that you can not make it to a general fact.

This is a complex issue. There is no correlation between nuclear size (which is what your question alludes to as opposed to cell size) and biological aggressiveness and there is in fact data showing that highly aneuploid cells proliferate more slowly than near diploid cells. However, the higher the degree of aneuploidy (which does correlate with nuclear size), the more likely the cancer cells are to behave abnormally including in their response to therapy. Also, it may be that the significance of high DNA content comes more from it being a marker of genomic instability, which should correlate with propensity for acquiring drug resistance, than from its direct phenotypic consequences

The power of malignancy can be measured by the ability of cancer cell to offer selective suitable environment for its growth. Normal there is some variation in cell size even in the same tissue, therefore the size of cells are not relevant for malignancy. How much cancer cells possess helpful genetic alteration how much it becomes malignant. Turn to CML where cells acquired Ph chromosome to become malignant cell where its size still not changed. In colon cancer, most cancerous cells showed larger size, while in basal cell carcinoma the cells are smaller in size. Hence we can conclude that cell size has no relation to malignant power.

there is no direct relationship between cell size and the agressivness of a tumor. However the most malignant ones tend to be the most promitive and therfore have small cells but off course there are many small cell cancers that are very indolent as for example CLL.

I do not think that there would be a correlation between the cell size and the malignant potential.

As pathologist I fully agree with Kate Lewis answer.

TUMOR SIZE AND MALIGNANCY / SURVIVAL: NON-LINEAR and NOT INVARIANT

I will below argue that although tumor size and outcome, often mediated by lymph-node involvement, are associated for many malignancies, the pattern of association is (1) not uniformly linear over all size values, and (2) is not invariable under all conditions. I will give some data supporting this from breast cancers, but I have found roughly comparable patterns across multiple malignancy types.

As Jennifer Wo and colleagues [1] at Dana-Farber have demonstrated in cases of extensive lymph node involvement, very small tumors may confer a more aggressive phenotype than larger tumors with the same degree of node-positivity. Thus small primary tumor size, defined as T1a tumors, when in the context of significant lymph node involvement (N2), appears to function as a surrogate for inherent, biologically aggressive disease and predict for higher breast cancer specific mortality (BCSM) compared with larger tumors with a comparable degree of nodal involvement. Specifically, their study found that women with smaller (T1a) tumors had a higher breast cancer–specific mortality (BCSM) than those with larger tumors (T1b), in the N2 context (the difference was not observed in patients with fewer or no lymph nodes involved (N1 or N0); indeed, no significant difference was found among patients with T1a tumors relative to patients with T1c or T2 tumors. Since however BCSM increases according to traditional expectations after a set threshold, with increasing tumor size conferring increased mortality, this further suggests the biphasic rather than wholly continuous nature of the tumor-to-BCSM relationship in certain circumstances.

These very small tumors may represent a subclass of breast cancers in which the metastatic progression, migration and invasion pathways are more dominant than primary tumor "self-seeding" pathways, and for this reason despite their size these hyper-malignant tumors (as this reviewer dubs them), do not, unlike other tumors, need cell accumulation and consequent growth to gain metastatic potential at an early, not late, acquisition event in cancer development as has been traditionally thought, via late-disseminating cells (in agreement at least in part with the recent commentary of Elizabeth Comen, Larry Norton and Joan Massagué [2] at MSK, and as the Canadian team of Steven narod and colleagues has concluded, collectively the evidence questions the long-held assumption that tumor size should always be a critical part of any treatment decision [3].

Zhi-Ming Shao's team in Shanghai [4] examined the relationship between large tumor size and breast cancer–specific mortality (BCSM), especially in a subset of patients who were (lymph) node-negative, using the large SEER database of 107,705 female patients, finding that using 21- to 30-mm tumors as the reference, the hazard ratio (HR) of breast cancer specific mortality increased with increasing tumor size until a peak at 41 to 50 mm (HR, 1.49), after which - namely > 5 cm. - increasing tumor size was unexpectedly related to decreasing hazard, with a nadir at 61 to 80 mm. The 61- to 80-mm tumors exhibited a significantly lower breast cancer specific mortality compared with the 41- to 50-mm and greater than 80-mm subgroups, a pattern not observed in patients with node-positive disease. Upon further analysis, it was found the survival advantage of 61- to 80-mm tumors in node-negative disease might be consequent to its low risk of distant metastasis. Hence we can conclude that a relatively larger tumor size without lymph node involvement might in fact be a surrogate for biologically indolent disease of regional/distant metastasis, and impressively, the breast cancer specific mortality associated with these larger 61- to 80-mm tumors was comparable with that of the reference standard of 21- to 30-mm tumors, but again these relationships between tumor size and survival were not observed in patients with node-positive disease.

In addition, Carlos Castaneda and colleagues [5] in Perú recently tracked the behaviour of molecular subtypes of breast cancer at different BC tumor sizes, finding that in general larger tumors of luminal A (hormone-positive), luminal B (endocrine with high proliferation (Ki-67) index) and basal/triple negative subtype are more likely to carry aggressive markers and to have worse outcomes than small ones, but this was not true of tumors of the HER2-positive subtype, where tumors behaved agreesively regardless of tumor size, large or small.

CLINICAL LESSONS

The disconnect between tumor size, lymph-node status and prognosis/survival in some subtypes of breast cancers and in some ranges of tumor size (61 - 80 mm) be secondary to (1) an underlying non-linear relationship between the number metastasis-initiating cells and the size of the cancer, and/or (2) that certain collections of small tumor cells might harbor a new class of hyper-malignant cells that are inherently ultra-aggressive and hence with a high likelihood to metastasize.

The clinical implications are considerable. First, this suggests that there is a class of patients with small hyper-malignant tumors that have intrinsically substantial risk of high mortality and so dictate early and aggressive therapeutic interventions. Second, this gives a new importance to early screening (like mammography and MRI) which may help avoid the development of distant metastasis and significantly reduced patient survival by very early stage tumor removal of such small hyper-malignant tumors. Third, the realization that the association between tumor size and survival is neither invariant nor consistently linear should provide a natural corrective in therapeutic decisions that traditionally failed to treat small tumors in general aggressively on the one hand, and treated all large tumors aggressively on the other despite the newly established facts that (1) some small tumors are hyper-malignant, and (2) some larger tumors are highly indolent, and as clinicians we need to begin to inform clinical treatment decisions by these new realizations.

References

1. Wo JY, Chen K, Neville BA, et al. Effect of very small tumor size on cancer-specific mortality in node-positive breast cancer. J Clin Oncol 2011 Jul 1; 29(19):2619-27.

2. Comen EA, Norton L, Massagué J. Breast cancer tumor size, nodal status, and prognosis: biology trumps anatomy. J Clin Oncol 2011 Jul 1; 29(19):2610-2.

3. Foulkes WD, Reis-Filho JS, Narod SA. Tumor size and survival in breast cancer--a reappraisal. Nat Rev Clin Oncol 2010 Jun; 7(6):348-53.

4. Yu KD, Jiang YZ, Chen S, et al. Effect of large tumor size on cancer-specific mortality in node-negative breast cancer. Mayo Clin Proc 2012; 87(12):1171-80.

5. Castaneda CA, Andrés E, Barcena C, Gómez HL, Cortés-Funés H, Ciruelos E. Behaviour of breast cancer molecular subtypes through tumour progression. Clin Transl Oncol 2012; 14(6):481-5

The malignant behavior depends upon many factors, a long list like the capacity of invasion (related to the expression of matrix degrading enzymes), the surface adhesion molecules of the cells allowing adhesion to the vessels and stroma, the secretion of immunosupressive factors, the hiding of surface receptors allowing the cells to escape from immune system etc. As you might expect, these abilities are not always correlated to aneuploidia. Also some of the most agressive tumors are named as "small cell tumors". One other factor that may be important is the percentage of tumor initiating cells or stem cells in a tumor cell population. The grading systems in histopathology are very frequently prepared according to a description of many tumor properties and searching these features to predict the patient outcome. This allows identification of many cellular, architectural and even tumor stromal factors as prognostic factors and so the grading systems evolve.

Cell heterogeneity is commonly seen among malignant cells. Different cell size cells can be detected in the same malignant tumour. Genetic component and other cytoplasmic activities within a cell may play an important factor to identify the cell cell size. When cancer cells contain tetraploid chromosome one would expect to find large cells due to multiplication of a full set of chromosomes. In solid tumours, different subclones can be seen and all those subclones show marked variation in cell size. Therefore it can be concluded that cell size shows no relation with malignant power.

I (KE) am working on significance of transient polyploidy of tumour cells for 15 years. I like how Christer Ericsson formulated the question. It is composed of three notions/assumptions and I shall comment each of them from the point of my experience (mostly experimental on tumour cell lines). (1) “Since there is a rough biological correlation between the amount of chromosomal material and cell size, one could expect those aneuploid cells with large duplications to be both large and highly malignant’. KE: That is true – they have larger potential, due to gene dosage, genome instability, which brings to aneuploidy and increases mutability, and the capability to induce stemness shown in 2010 by our group. (2) “But since small cells may have become small during the evolution of the tumor, exactly since they have shed all "unnecessary" function in order to focus on growing and dividing, THEY may be expected to be the most malignant’. KE: This is a fair kit. We have published two experimental articles, in 2000 and 2011 and one review in 2012 to show that just this is occurring as a result of the prolonged process, both natural and after DNA damaging treatments, when it takes about two weeks. A polyploidy mother gives ‘birth’ to para-diploid small rejuvenated daughters casting of the unwanted material (cytoplasmic and nuclear’). (3)” On the third hand since it is NOT part of the clinical pathologists grading system of cancers to rate the cell size, one could expect that NORMAL sized cells would include the most malignant. So, which is it?’ KE: First, clinical cytologists, at least some in Germany, France and Norway do use the measurements of the DNA contents for diagnosing some malignancies. F.ex., if you find even one octaploid cell in a cervical smear you can suspect carcinoma in situ rather certainly. To measure the DNA content of paraffin slides is tricky. Second, the survival rate in the process described above is very low (10-4-10-6 cells), because most giant cells are transient, instable and undergo death or senescence; the process of recovery through reversible polyploidy is so slow that the pathologists usually do not notice it and therefore, yes, normal-sized cells are malignant, however in fact there is a conveyer from polyploidy to the normal-sized malignant cells. This is not only the experience of my group in collaboration with English group of prof. M.S. Cragg. This data on acquiring of stemness (immortality) and resistence through REVERSIBLE polyploidy (not to be mixed up with the permanent aneuploidy) have been recently confirmed by several groups on lymphoma, breast carcinoma, NSCLC, osteosarcoma, colon cancer, on cell lines and some, on clinical material. For the materials, please, see my site in the ResearchGate (Jekaterina Erenpreisa), where practically all our experimental articles and reviews with references on the other colleagues work can be found. We have just submitted the latest review on the association of reversible polyploidy with stemness and cell senescence (and senescent cells are also large and often polyploidy). As soon as it is accepted, it will appear online and I shall attach it at the ResearchGate.

As Kate Lewis mentioned,it is not possible to generalize a link between the cell size and tumor behavior. The nucleus of malignant cell is often very much enlarged with out increasing the amount of cytoplasm,which often differentiate malignant cells from benign cells. This is also not common in all malignancies, for example papillary carcinoma of the thyroid,some tumors of the salivary gland etc.

yes I too agree with Kate Lewis. But I think studying of interactions in the signalling pathway will be helpful.

Cancer cells contain usually mitotic figures that may confer big nuclei. These figures reflect intense cellular proliferation. Furthermore, we know that small cell lung cancer grows rapidly and is more aggressive than non small one. It is just an example and each type of cancer has its specific aspects, so i think that we can not generalyze, because the behavior of a type of cancer is not only dependent on a given morphology, it is rather the result of many other parameters, such as anatomic localization, the nature and the state of tumor infiltrating immune cells, stromal cells, ability to repair DNA and mostly the interaction of all the micro-environmental factors of the tumor that determine its fate and consequently the prognosis of the disease.

There is lots of discussion on this concept but I want to add some more information. I fully agree with the previous comments regarding to the cell size and aggressiveness of a tumor cells. Size of tumor cell does not matter for aggressiveness but the amount of altered genome and the tumor cells’ outside microenvironment which includes growth factors, hormones etc. contributing to the aggressiveness of tumor cell and more importantly there are some metastatic protein expressed by the tumor cell which is expressed by some tumor cells more and by other less, is the main contributing factor in the decision of aggressiveness of tumor cells. Also sometimes aggressiveness is also enhanced by the local environment generated by the other disorders/diseases such as Diabetes, Obesity etc.

Well friends, a lot of confusion and contradiction! Let us try to simplify. Anaplasia, characterized by nuclear and cellular pleomorphism, macro- and aniso-nucleosis and -nucleolosis, coarse chromatin, increased mitosis, nuclear hyperchromasia, heteroploidy, aneuploidy, polyploidy, is the key feature that differentiates cancer cells from normal ones.The best way to recognize these features is cytology rather than histology. Cancer cells are abnormal and they appear and behave abnormally and that is how, we routinely visualize them under the microscope and by other means. In most cancers, usually one lineage of cells is predominant over others and moreover overcrowding of cells of that particular lineage is seen.Within this lineage, large, small and normal-like cells are found, that is what is termed pleomorphism. Now let us look at small, medium and large cancer cell sub-types as discussed in detail previously, they all indicate the origins or lineage of that cancer. For example, squamous cell carcinoma has mixed cell populations because it arises and differentiates, though abnormally into most stratas of stratified Squamous epithelium. Contrarily, the basal cell tumor has smaller cells as it arises from basal cells. In the same manner, small cell cancer of the lung or papillary thyroid carcinoma arises from basal and small cells and thus mainly has smaller or normally sized cells, respectively. Same holds true for lymphoma. In nutshell, cell size has no correlation to malignancy, although anaplasia holds the key, besides several other factors, including tissue microenvironment.

Cell size is independent form malignacy. Malignancy depends on the potential of the proliferative mutation. Central mutations, that allow replication at a central spot in the biochemical cascade, as the ALL-1 transformation, are highly malignant. Those cells shortcut all G1 or G0 phases of the cell cycle, and immediately replicate their DNA after mitosis. No regualation is possible, the cells grow fastest. Peripheral mutations, that allow replication in the periphery of proliferative genes, show some G1 phase, as replication initiation takes more time. Those cells grow slower and the immune system could catch their antigenicity. Cancer Genet Cytogenet. 1997 Sep;97(2):143-54 Proliferative genes induce somatic pairing defects in Drosophila melanogaster and allow replication.

Polyploidy and, particularly para-triploidy are prognostically unfavourable as known for decades from numerous studies on cancer patients of different cell types. The cause may be that reversible polyploidy induced by DNA damage, inherent or after genotoxic treatments is associated with de-differentiation to embryonal stemness and release of EMT-potent descendants. The embryonal stemness signature was reported also in vivo for the most aggressive tumours. Please, see our review.

Taken the cancer stem/initiating cells into account, a hierachy is expected to take place in the tissue. If it is true, then differently-sized cells are supposed to present in the same tissue (cancer stem cells--progenitors---differentiated ones). Possibly, the malignant potential of groups is different from each other although they may have different size (the stem-like cell being most aggresive while the differentiated one being least aggresive?). To me, the stem-like cells are smaller than the others. If it is really true, then i would say that the smaller, the more malignant! So, Christer, i love your point. The pathologists should look for the stem-like cells to better predict the prognosis as these cancer stem/initiating cells are directly related with prognosis. Good point!  

Taking analogy from plant and yeast hybrids that grow faster than both parents. From an evolutionary point of view such multiplicative phenotypes (such as grain yield and cancer progression) could be the result of trade-offs that are released (in hybrids and cancer cells). There may be a "price" for that, but cancerous cells are not healthy cells that keeps resonance with rest of body. So, perhaps worth thinking of that as mechanisms that breaks this negative relationship between cell size and growth, and able to achieve both? Plant hybrid manage to maintain robustness, or production of total seed mass, because of a multiplicative effect on seed number and size, two traits that are normally in negative correlation. But, first generation hybrids are not normal- they are artificially made by breeders, as cancer cells obtain non-normal lineage. In practice, it suggest that one may need to look for mutations that break trade offs.  

The size of the tumor cell generally depend on its linage. Most of the aggressive tumors show highly pleomorphic large malignant cells for example anaplastic carcinomas of lung thyroid etc.In breast also highly  pleomorphic malignant cells predict poor out come.However in small cell carcinomas the size of the cells is independent of the grade of the tumor. The size of the tumor cell indicate ploidy status; polyploidy is well known for its prognostically unfavorable nature. In general, the tumor cell size cannot be taken as a factor indicating the grade of malignancy

Cancer cells and their malignant power depend on what is going on inside cancer cells to confer better condition to become more malignant as to protect them self from the immune system. Usually the size of the cell depend on the size of the nucleus, much chromatin and genetic components mean more large nucleus and in turn large cell size. genetic instability is one of the most important characteristic feature of cancer cells but that doesn't mean that large nucleus represents more malignant power.The malignancy depends on what happens inside nucleus to make the cell strong against the microinvironment enemies.

Article Breast Cancer Tumor Size, Nodal Status, and Prognosis: Biolo...