• genetics;
  • protective genetic alterations;
  • long-term survivors;
  • neoplastic disease


  1. Top of page
  2. Abstract
  3. Studies in Patients with Long-Term Survival
  4. Studies in Individuals Potentially Protected from Developing Malignant Disease

The investigation of genetic alterations that may be related to the prognosis of patients with malignant disease has become a frequently used strategy in recent years. Although some conclusions have been reached in certain studies, the complexity and the multifactorial nature of most neoplastic diseases makes it difficult to identify clinically relevant information, and the results of some studies have been of borderline significance or have been conflicting. In contrast, the identification and the study of patients or families with very characteristic phenotypes have yielded outstanding results in the identification of the genetic characteristics underlying such phenotypes. Although, in most cases, the individuals who are selected for these types of studies are characterized by a negative phenotype (i.e., individuals who are at increased risk for developing a specific disease), a few studies have been directed toward individuals with phenotypes that imply an unusually good prognosis (i.e., individuals who present with a decreased risk for developing specific diseases despite an important exposure to well-known risk factors). Therefore, it seems logical to develop this strategy further as a valid methodology for the study of other diseases, such as cancer. The study of individuals with phenotypes that imply an extremely good prognosis, such as long-term survivors of theoretically incurable malignancies or individuals who seem to be protected against a certain neoplastic disorder despite having a markedly increased risk for its development, may unveil genetic alterations that explain such characteristic phenotypes and may provide potentially useful therapeutic targets against these diseases. Cancer 2002;95:1605–10. © 2002 American Cancer Society.

DOI 10.1002/cncr.10877

During recent years, considerable numbers of studies have correlated the presence of genetic alterations in patients with malignant disease or in tumors with their prognosis or with the efficacy of several therapeutic agents, following the hypothesis that different individuals or tumors may harbor diverse genetic alterations that may correlate with their clinical outcome. Although the validity of such an approach is unquestionable, one potential drawback is the possibility that, even if the genetic alterations studied are related to the outcome of the patient, such alterations may not be the only factor or the main prognostic factor that determines patient outcome. Instead, because cancer is a complex and multifactorial disease, the prognosis of patients is the result of the interaction of many intrinsic and extrinsic factors. Consequently, the identification of genetic alterations that are associated with a significantly different prognosis becomes a difficult task, because most of such alterations have a reduced penetrance,1 compared with diseases that are caused by fully penetrant genetic alterations, such as cystic fibrosis, in which the presence of the genetic alteration is linked inexorably with the development of the disease.

A potentially more efficient strategy for isolating the genetic features associated with characteristic outcomes may be to concentrate studies in few individuals with extremely differentiated phenotypes, which have a greater chance of carrying the characteristic genotypes responsible for the differences in prognosis that are seen in the clinic, rather than to perform studies in the whole population. Such an approach is far from new in medicine and has been employed successfully, even in oncology. Indeed, the study of individuals who are affected by multiple tumor syndromes and/or individuals with a markedly increased familial risk of developing malignant disease has led to the discovery of many genetic alterations that explain such situations. A few well-known examples are the identification of p53 germ line mutations in patients with the Li–Fraumeni syndrome,2 which was described through the identification of an excess in the risk of death by rhabdomyosarcoma in siblings;3, 4 the finding that patients with hereditary retinoblastoma present with an inactivation of both copies of the retinoblastoma tumor-suppressor gene,5, 6 as wisely predicted by Knudson;7 or the detection of mutations in BRCA-1, a gene that was identified in families with early-onset breast carcinoma.8 Nevertheless, a phenotype may be considered characteristic based on considerations other than an increased risk of developing a specific malignant disease, and yet it can be explained by genetic abnormalities. The identification of a complete deficiency of dihydropyirimidine dehydrogenase activity in peripheral blood mononuclear cells in a patient who developed severe toxicity after receiving 5-fluorouracil administration is a good example:9 Even though, in the initial report, the molecular explanation for this deficiency was not identified, several subsequent studies have demonstrated that diverse genetic alterations are associated with this deficiency of enzymatic activity.10–12

These examples show that the yield from studying the causes of a very characteristic phenotype is very high, because only a few patients (or even just one patient) must be studied to identify the factors that are linked strongly to the phenotype. This is logical, because the only hypothesis tested is whether the observed phenotype characteristics—which already have been identified and are so unique that it is highly unlikely they are due to chance—are related to a determined cause. In contrast, the hypothesis that a specific genetic alteration confers a determined prognosis, either improved or worsened, is more uncertain, because it implies two hypothesis: 1) that the patients studied have a characteristic prognosis and 2) that the genetic alteration being studied explains such a difference. Thus, it is more difficult to reach a reliable conclusion.

Studies in Patients with Long-Term Survival

  1. Top of page
  2. Abstract
  3. Studies in Patients with Long-Term Survival
  4. Studies in Individuals Potentially Protected from Developing Malignant Disease

An alternative approach using the same methodology would be to analyze the rare individuals who are unexplained long-term survivors of neoplasms that are supposed to be incurable and to identify the reasons that may explain such prolonged survival. A limited number of case reports on long-term survivors of diseases such as gastric carcinoma,13–15 colon carcinoma,16 pancreatic carcinoma,17 or melanoma18 in apparently incurable form can be found in the literature. Interestingly, the strategy of studying groups of patients with malignant disease who are long-term survivors has been employed in the past in some trials, for example, in patients with nonsmall cell lung carcinoma,19–21 although those studies focused on the identification of clinical prognostic factors rather than genetic factors. If the existence of patients with malignant disease who have an unusually prolonged survival is confirmed, then studying such patients may unveil the causes that explain their survival. These causes may include tumor factors (apoptotic pathways, drug resistance, etc.) or host factors (drug metabolism, immunologic response, etc.) in addition to external factors. The finding of more than one of these patients within the same family would argue strongly in favor of such a hypothesis. Evidently, the diagnosis of the individuals who are selected for these kinds of studies has to be based on solid histologic evidence, because it is possible that some patients may have been diagnosed erroneously, which would provide an underlying explanation for their cure.

Studies in Individuals Potentially Protected from Developing Malignant Disease

  1. Top of page
  2. Abstract
  3. Studies in Patients with Long-Term Survival
  4. Studies in Individuals Potentially Protected from Developing Malignant Disease

Another step ahead in the same direction would be to hypothesize that, because the risk of developing a malignancy is not uniform, just as some individuals present with an increased risk, others may have lower risk than would be expected for their environment and habits. These individuals would represent the left tail of a Gaussian distribution showing the risk of developing cancer. If these individuals exist, then their identification and the study of the causes of such protection may increase our knowledge about malignant disease and also may yield potentially useful treatments against it. Furthermore, we must observe that malignancy is a frequent and ancient disease—probably inherent to life itself—and, thus, it is possible that genetic alterations that confer partial protection or even total protection against certain neoplastic disorders may have developed during evolution. For example, such protection from malignant disease may be related to mechanisms of DNA repair, activation of apoptotic pathways, immunologic responses, or other factors.

Again, this approach is not new in medicine, and two intriguing examples can be used to illustrate successful strategies for identifying genetic alterations that confer protection against a disease. Possibly the most interesting example is studying the genotype of patients with a unique, very characteristic phenotype. An outstanding example is the identification of a deletion in the gene encoding the chemokine coreceptor CCR-5, which confers to homozygotic individuals complete protection against infection by certain strains of the human immunodeficiency virus (HIV).22, 23 It has been shown that at least one other heterozygotic CCR-5 mutation, when associated with the same deletion in the other allele, produced the same effect.24 Because the CCR-5 mutations had not been associated with any abnormalities, an astute observation of the fact that some individuals highly exposed to HIV never developed the infection was required to identify them.25 Secondary to this observation, CCR-5 has become a relevant target in the investigation of HIV infection.

A second strategy for identifying protective genetic profiles has been to study the expression of gene polymorphisms that supposedly are related to the incidence of a determined disease in case-control studies. A recent example is the relation between certain factor VII genotypes and the risk of myocardial infarction.26 It has been suggested that high plasma levels of coagulation factor VII are correlated with the risk of death due to coronary artery disease, and polymorphisms in the factor VII gene are associated with variations in levels of factor VII. Therefore, these polymorphisms were studied in 311 individuals with severe, angiographically documented coronary atherosclerosis, 175 of whom had a history of previous myocardial infarction. Among patients with no history of previous myocardial infarction, there were significantly greater numbers of patients with determined genotypes compared with the numbers of patients with myocardial infarction, thus suggesting a protective action for patients with such genotypes. These types of studies must be interpreted with criticism, because they are subject to a potentially high risk of bias, as discussed in the review by Gambaro et al.27 However, this example is of particular interest, because the individuals identified who were protected against ischemic disease were asymptomatic and had no history of myocardial infarction, although they were at high risk for developing it—documented coronary atherosclerosis—and were not just normal, healthy individuals (although the study included a control group of healthy participants). Thus, there was a greater opportunity to find true protective factors rather than simply the absence of disease, which would be expected in a group of healthy individuals with a normal population risk.

The bottom line for both examples is that the study of individuals with a lower than normal risk of developing a disease may unveil protective host factors for that disease, just as the study of individuals with an elevated risk of developing malignant disease may lead to the discovery of cancer-related genetic alterations. The proof of the existence of cancer-protective genetic alterations comes from the description of the resistance to develop breast carcinoma induced by the neu and ras oncogenes, as observed in mice lacking the cyclin D1 gene.28 Although the cyclin D1 inactivation in these mice was induced artificially,29, 30 the relatively mild effects that it produced over adult mice physiology raises the question of whether this same ablation may have taken place through spontaneous mechanisms (e.g., spontaneous mutations). If that is the case, then the individuals who bear such mutations would be protected against the development of breast carcinoma, even though the protection would be only partial, because the cyclin D1-deficient mice were not protected against breast carcinoma that was induced through other oncogenic pathways, such as c-myc or Wnt-1.

Relevant numbers of human polymorphisms that supposedly are related to some degree of protection against developing malignant disease have been described. Certain polymorphisms of enzymes, like myeloperoxidase,31–34 NAD(P)H:quinone reductase,35 or microsomal epoxide hydrolase,36 which are involved in the metabolism of determined carcinogens, have been related to some degree of protection against developing lung carcinoma or colorectal carcinoma37 in some population subsets and/or ethnic groups, even though other studies, such as that of microsomal epoxide hydrolase, have not shown any association38 or even showed an inverse relation.39 Polymorphisms in the methylenetetrahydrofolate reductase gene have been associated with a decreased risk of developing acute lymphocytic leukemia40 and colorectal carcinoma41, 42 in determined population groups. Determined genotypes of some cytochrome P450 enzymes, such as CYP1A1 or CYP2D6, also have been correlated with a decreased risk of lung carcinoma, although meta-analyses have failed to confirm this observation43, 44 or have described only a small protective effect with a nonappreciable relation to individual susceptibility.45 Two common polymorphisms of the p21WAF1/Cip1 gene have been associated with a potential protective role against ovarian carcinoma.46 Some genotypes of glutathione S-transferase M1 also have been related to the risk of lung carcinoma and other aerodigestive tract malignancies; although, again, a meta-analysis failed to confirm such results.47 Finally, some histocompatibility antigen alleles have been linked in case-control studies with a decreased susceptibility to lung carcinoma,48 melanoma,49 and renal cell carcinoma,50 and even homozygotic women for determined polymorphic alleles of the BRCA-1 gene have been associated with a decreased risk of breast carcinoma.51

The ambiguous and clinically not very relevant results of some of these studies may be explained by flaws in methodology, as detailed elsewhere,27 but also may be related in part to an inadequate selection of the study populations. Compared with the former studies, in which individuals were selected by a very characteristic phenotype—a definite protection against developing HIV or myocardial infarction—the latter studies compared the risk of patients who have developed disease with a control group formed by normal control participants. In these individuals, the risk of developing disease probably was neither increased nor decreased—perhaps the only exception was that some groups were formed by smokers—therefore, it was unlikely that clinically relevant information would be discovered. A potentially more efficient approach may be to study individuals with a markedly reduced familial or individual risk of developing malignant disease. Families with a very low or ideally null incidence of malignant disease over several generations, perhaps despite crossing with high-risk families, may have a reduced familial risk. Individuals who do not develop malignant disease despite important exposure to well-known intrinsic factors (such as a potential patient with familial adenomatous polyposis developing malignant disease significantly later than would be expected or not developing it at all) or extrinsic factors (such as heavy exposure to radiation) may have a reduced individual risk. Combinations of these strategies or different strategies also may be pursued, keeping in mind that the possibility of yielding positive results will be related directly to the discrepancy between the risk of developing malignant disease and the actual phenotype. If such families or individuals exist, then it would be naïve to attribute their characteristic phenotype to chance, at least until other causes have been ruled out.

In summary, the study of individuals with very characteristic phenotypes has been very useful for describing the mechanisms underlying them. This has been confirmed in individuals who are at high risk for specific diseases as well as in individuals who seem to be protected against certain diseases. Therefore, it is logical to pursue this strategy as a valid methodology for the study of other diseases, such as cancer. The creation of data bases compiling clinical and environmental information as well as adequate samples from tumor tissue and/or normal tissue from individuals who are long-term survivors of theoretically incurable malignancies and from individuals who seem to be protected against certain neoplastic disorders, along with the study of such data, may help to increase our current knowledge of malignant diseases and to discover new therapeutic strategies against them. Even in an age of computer-aided molecular biology, observation should remain a better way than speculation to generate valid hypotheses.


  1. Top of page
  2. Abstract
  3. Studies in Patients with Long-Term Survival
  4. Studies in Individuals Potentially Protected from Developing Malignant Disease
  • 1
    Todd JA. Interpretation of results from genetic studies of multifactorial diseases. Lancet. 1999; 354: 1516000.
  • 2
    Malkin D, Li FP, Strong LC, et al. Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms. Science. 1990; 250: 12331238.
  • 3
    Miller RW. Deaths from childhood cancer in sibs. N Engl J Med. 1968; 279: 122126.
  • 4
    Li FP, Fraumeni JF Jr. Rhabdomyosarcoma in children: epidemiologic study and identification of a familial cancer syndrome. J Natl Cancer Inst. 1969; 43: 13651373.
  • 5
    Benedict WF, Murphree AL, Banerjee A, Spina CA, Sparkes MC, Sparkes RS. Patient with 13 chromosome deletion: evidence that the retinoblastoma gene is a recessive cancer gene. Science. 1983; 219: 973975.
  • 6
    Cavenee WK, Dryja TP, Phillips RA, et al. Expression of recessive alleles by chromosomal mechanisms in retinoblastoma. Nature. 1983; 305: 779784.
  • 7
    Knudson AG Jr. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA. 1971; 68: 820823.
  • 8
    Hall JM, Lee MK, Newman B, et al. Linkage of early-onset familial breast cancer to chromosome 17q21. Science. 1990; 250: 16841689.
  • 9
    Diasio RB, Beavers TL, Carpenter JT. Familial deficiency of dihydropyrimidine dehydrogenase. Biochemical basis for familial pyrimidinemia and severe 5-fluorouracil-induced toxicity. J Clin Invest. 1988; 81: 4751.
  • 10
    Van Kuilenburg AB, Vreken P, Abeling NG, et al. Genotype and phenotype in patients with dihydropyrimidine dehydrogenase deficiency. Hum Genet. 1999; 104: 19.
  • 11
    Van Kuilenburg AB, Haasjes J, Richel DJ, et al. Clinical implications of dihydropyrimidine dehydrogenase (DPD) deficiency in patients with severe 5-fluorouracil-associated toxicity: identification of new mutations in the DPD gene. Clin Cancer Res. 2000; 6: 47054712.
  • 12
    Raida M, Schwabe W, Hausler P, et al. Prevalence of a common point mutation in the dihydropyrimidine dehydrogenase (DPD) gene within the 5′-splice donor site of intron 14 in patients with severe 5-fluorouracil (5-FU)-related toxicity compared with controls. Clin Cancer Res. 2001; 7: 28322839.
  • 13
    Miyaji M, Ogoshi K, Kajiura Y, et al. [A case of advanced gastric cancer with liver metastasis with no recurrence and long survival.] Gan To Kagaku Ryoho. 1996; 23: 915918.
  • 14
    Mura T, Bandou H, Nomura T, Watanabe K, Kobayashi T. [A resected case of advanced gastric cancer with complete remission of liver metastasis by chronic daily administration of oral etoposide and UFT.] Gan To Kagaku Ryoho. 1992; 19: 10711074.
  • 15
    Sasaki Y, Sasaki J, Ito T, Nehashi Y. [Complete response in a case of unresectable gastric cancer with a combination of tegafur, 5-fluorouracil and mitomycin C.] Gan To Kagaku Ryoho. 1988; 15: 27932795.
  • 16
    Mukai M, Tokunaga N, Yasuda S, et al. Long-term survival after immunochemotherapy for juvenile colon cancer with peritoneal dissemination: a case report. Oncol Rep. 2000; 7: 13431347.
  • 17
    Silberstein E, Walfisch S, Lupu L, Sztarkier I. Twelve-year survival after the diagnosis of locally advanced carcinoma of the pancreas: a case report. J Surg Oncol. 2000; 75: 142145.
  • 18
    Dutcher JP, Wiernik PH. Long-term survival of a patient with multiple myeloma—a cure? A case report. Cancer. 1984; 53: 20692072.
  • 19
    Finkelstein DM, Ettinger DS, Ruckdeschel JC. Long-term survivors in metastatic non-small-cell lung cancer: an Eastern Cooperative Oncology Group Study. J Clin Oncol. 1986; 4: 702709.
  • 20
    Satoh H, Ishikawa H, Yamashita YT, et al. Analysis of long-term survivors after platinum containing chemotherapy in advanced non-small cell lung cancer. Anticancer Res. 1998; 18: 12951298.
  • 21
    Sculier JP, Paesmans M, Libert P, et al. Long-term survival after chemotherapy containing platinum derivatives in patients with advanced unresectable non-small cell lung cancer. European Lung Cancer Working Party. Eur J Cancer. 1994; 30: 13421347.
  • 22
    Liu R, Paxton WA, Choe S, et al. Homozygous defect in HIV-1 correceptor accounts for resistance of some multiply-exposed individuals to HIV-1 infection. Cell. 1996; 86: 367377.
  • 23
    Samson M, Libert F, Doranz BJ, et al. Resistance to HIV-1 infection in Caucasian individuals bearing mutant alleles of the CCR-5 chemokine receptor gene. Nature. 1996; 382: 722725.
  • 24
    Quillent C, Oberlin E, Braun J, et al. HIV-1-resistance phenotype conferred by combination of two separate inherited mutations of CCR5 gene. Lancet. 1998; 351: 1418.
  • 25
    Rowland-Jones S, Sutton J, Ariyoshi K, et al. HIV-specific cytotoxic T-cells in HIV-exposed but uninfected Gambian women. Nat Med. 1995; 1: 5964.
  • 26
    Girelli D, Russo C, Ferraresi P, et al. Polymorphisms in the factor VII gene and the risk of myocardial infarction in patients with coronary artery disease. N Engl J Med. 2000; 343: 774780.
  • 27
    Gambaro G, Anglani F, D'Angelo A. Association studies of genetic polymorphisms and complex disease. Lancet. 2000; 355: 308311.
  • 28
    Yu Q, Geng Y, Sicinsky P. Specific protection against breast cancer by cyclin D1 ablation. Nature. 2001; 411: 10171021.
  • 29
    Sicinski P, Donaher JL, Parker SB, et al. Cyclin D1 provides a link between development and oncogenesis in the retina and breast. Cell. 1995; 82: 621630.
  • 30
    Fantl V, Stamp G, Andrews A, Rosewell I, Dickson C. Mice lacking cyclin D1 are small and show defects in eye and mammary gland development. Genes Dev. 1995; 9: 23642372.
  • 31
    London SJ, Lehman TA, Taylor JA. Myeloperoxidase genetic polymorphism and lung cancer risk. Cancer Res. 1997; 57: 50015003.
  • 32
    Schabath MB, Spitz MR, Zhang X, Delclos GL, Wu X. Genetic variants of myeloperoxidase and lung cancer risk. Carcinogenesis. 2000; 21: 11631166.
  • 33
    Le Marchand L, Seifried A, Lum A, Wilkens LR. Association of the myeloperoxidase-463 G [RIGHTWARDS ARROW] A polymorphism with lung cancer risk. Cancer Epidemiol Biomarkers Prevent. 2000; 9: 181184.
  • 34
    Cascorbi I, Henning S, Brockmoller J, et al. Substantially reduced risk of cancer of the aerodigestive tract in subjects with variant-463A of the myeloperoxidase gene. Cancer Res. 2000; 60: 644649.
  • 35
    Chen H, Lum A, Seifried A, Wilkens LR, Le Marchand L. Association of the NAD(P)H:quinone oxidoreductase 609 C [RIGHTWARDS ARROW] T polymorphism with a decreased lung cancer risk. Cancer Res. 1999; 59: 30453048.
  • 36
    London SJ, Smart J, Daly AK. Lung cancer risk in relation to genetic polymorphisms of microsomal epoxide hydrolase among African-Americans and Caucasians in Los Angeles County. Lung Cancer. 2000; 28: 147155.
  • 37
    Harth V, Donat S, Ko Y, Abel J, Vetter H, Bruning T. NAD(P)H quinone oxidoreductase 1 codon 609 polymorphism and its association to colorectal cancer. Arch Toxicol. 2000; 73: 528531.
  • 38
    Smith CA, Harrison DJ. Association between polymorphism in gene for microsomal epoxide hydrolase and susceptibility to emphysema. Lancet. 1997; 350: 630633.
  • 39
    Benhamou S, Reinikainen M, Bouchardy C, Dayer P, Hirvonen A. Association between lung cancer and microsomal epoxide hydrolase genotypes. Cancer Res. 1998; 58: 52915293.
  • 40
    Skibola CF, Smith MT, Kane E, et al. Polymorphisms in the methylenetetrahydrofolate reductase gene are associated with susceptibility to acute leukemia in adults. Proc Natl Acad Sci USA. 1999; 96: 1281012815.
  • 41
    Chen J, Giovanucci E, Kelsey K, et al. A methylenetetrahydrofolate reductase polymorphism and the risk of colorectal cancer. Cancer Res. 1996; 56: 48624864.
  • 42
    Ma J, Stampfer MJ, Giovanucci E, et al. Methylenetetrahydrofolate reductase polymorphism, dietary interactions, and risk of colorectal cancer. Cancer Res. 1997; 57: 10981102.
  • 43
    Houlston RS. CYP1A1 polymorphisms and lung cancer risk: a meta-analysis. Pharmacogenetics. 2000; 10: 105114.
  • 44
    Christensen PM, Gotzsche PC, Brosen K. The sparteine/debrisoquine (CYP2D6) oxidation polymorphism and the risk of lung cancer: a meta-analysis. Eur J Clin Pharmacol. 1997; 51: 389393.
  • 45
    Rostami-Hodjegan A, Lennard MS, Woods HF, Tucker GT. Meta-analysis of studies of the CYP2D6 polymorphism in relation to lung cancer and Parkinson's disease. Pharmacogenetics. 1998; 8: 227238.
  • 46
    Milner BJ, Brown I, Gabra H, Kitchener HC, Parkin DE, Haites NE. A protective role for common p21WAF1/Cip1 polymorphisms in human ovarian cancer. Int J Oncol. 1999; 15: 117119.
  • 47
    Houlston RS. Glutathione S-transferase M1 status and lung cancer risk: a meta-analysis. Cancer Epidemiol Biomarkers Prevent. 1999; 8: 675682.
  • 48
    Tokumoto H. Analysis of HLA-DRB1-related alleles in Japanese patients with lung cancer—relationship to genetic susceptibility and resistance to lung cancer. J Cancer Res Clin Oncol. 1998; 124: 511516.
  • 49
    Ichimiya M, Muto M, Hamamoto Y, Ohmura A, Tateno H, Asagami C. Putative linkage between HLA class I polymorphism and the susceptibility to malignant melanoma. Australas J Dermatol. 1996; 37 (Suppl 1): S39.
  • 50
    Ozdemir E, Kakehi Y, Nakamura E, et al. HLA-DRB1*0101 and *0405 as protective alleles in Japanese patients with renal cell carcinoma. Cancer Res. 1997; 57: 742746.
  • 51
    Dunning AM, Chiano M, Smith NR, et al. Common BRCA1 variants and susceptibility to breast and ovarian cancer in the general population. Hum Mol Genet. 1997; 6: 285289.