In 2015 alone, the American Cancer Society estimates that 1,658,370 new cases of cancer will be diagnosed in the United States of which 8 percent will be colorectal cancer.[1] Each one of these new 1,658,370 cancer cases will be different from the others due to cancer's ability to be highly unique to its host and oncologists will have to personalize a treatment plan for each case. Although treatment options have improved vastly improved over the past century, 564,800 people are expected to die from cancer this year in the United States—more than 1,500 people per day.[2] Worldwide, 7.6 million people die each year from cancer, 4 million of which die prematurely (30 to 69 years of age).[3] Colorectal cancer is the third most frequently diagnosed cancer and the third deadliest cancer in both men and women. Although it can usually be detected early and quickly eradicated through screening, not all people are lucky enough to be diagnosed when the cancer is in its early stages. Of these unlucky cases is Jennie whose colorectal cancer was detected after she began showing symptoms. Jennie had a specific type of cancer referred to as hereditary nonpolyposis colorectal cancer (HNPCC) or Lynch syndrome caused by a mutation in her DNA mismatch repair system. Although this type of hereditary cancer made treating Jennie more difficult, her doctors were successful in keeping the cancer at bay and Jennie has been cancer free for almost two years.
PATIENT: Jennie AGE: 90 years GENDER: Female ETHNICITY: Chinese American

Diagnosis of Colorectal Cancer and Lynch Syndrome



FAMILY HISTORY: Jennie’s mother had uterine and rectal cancer and died at age 59; her father died of heart failure at age 78. Sister #1 had a sarcoma on her left knee, colon cancer, a mastectomy on the left breast, and pancreatic cancer that took her life at age 66. Sister #3 had ovarian and colon cancer and died at age 76. Sister #4 had ovarian and cecal colon cancer and died at age 68. Sister #5 had ovarian, bladder, and pancreatic cancer and died at age 67. Sister #6 currently lives with colon cancer and diabetes. Lastly, her brother had skin, colon, and prostate cancer and died at age 78.
CDR415500-750.jpg
Figure 1. Colon polyps.

HISTORY OF PRESENT ILLNESS: The patient presented to medical attention after a routine colonoscopy showed polyps in March of 1975. The polyps were removed and sent to pathology where results showed colon cancer; Jennie was treated with radiation and chemotherapy. Relapse ensued in late 1989 and the patient was again treated with radiation and chemotherapy until remission. White Cross Hospital admitted Jennie late on September 28, 2012 as she was complaining of a low-grade fever, low appetite, and fatigue. Initial diagnosis was pneumonia and antibiotics were administered. During her hospital stay, the patient showed blood in the stool and low-grade microcytic anemia, characterized by small red blood cells or microcytes, while complaining of lower abdominal cramping and pressure in the bowel. As symptoms were consistent with colorectal cancer, she underwent a colonoscopy on September 30, 2012 by Dr. Smith. Pathology confirmed stage IIIC colorectal cancer. Of note, the patient was previously diagnosed with bladder cancer, non-Hodgkin’s lymphoma, and skin cancer. The recurring cancer compiled with the patient’s familial history of cancer encouraged Dr. Smith to send Jennie to the UCLA Cancer Genetics Program where she underwent a blood test for indicators of specific mutations linked to cancer predisposition on November 6, 2012. Test results indicated positive for a deleterious mutation in MSH2 sequencing.
This diagnosis confirmed the presence of Lynch syndrome, also known hereditary non-polyposis colorectal cancer (HNPCC), and was the result of the in-frame deletion of amino acids 265-314 in the MSH2 gene.

Hallmarks of Cancer: The Molecular Basis of MSH2


hanahan2011.jpg
Figure 2. Hanahan and Weinberg's Hallmarks of Cancer.




In recent years, Hanahan and Weinberg have published an article entitled Hallmarks of Cancer: The Next Generation. In it, the two argue that there are six hallmarks of cancer—“distinctive and complementary capabilities that enable tumor growth and metastatic dissemination” (Figure 2).[4] Mutations in MSH2 create an environment where at least two of these hallmarks, evading growth suppressors and resisting cell death, are already present and as a result, cancer more easily develops.




mismatch repair 1.jpg
Figure 3. Nucleotide illustration of MSI.



MutS homolog 2 (MSH2) is a gene that codes for mismatch repair (MMR) proteins and is one of the two most common genes involved with high levels of microsatellite instability (MSI). In Jennie’s case, an inherited deleterious mutation led to the inactivation of the MSH2 MMR gene and thus, “the accumulation of errors within microsatellites” (also known as short tandem repeats or STRs—repeating sequences of four to five base pairs of DNA) or MSI. MSI caused the increased likelihood of tumor formation and predisposed Jennie to cancer.[5]


nri858-i1.gif
Figure 4. DNA mismatch repair.
In MSH2 deleterious mutations, a section of the gene encoding MSH2 is absent and the mismatch repair protein cannot function. DNA mismatch repair (MMR) is responsible for recognizing and correcting any incorrectly paired nucleotides that arise from errors during DNA replication or from DNA damage. Years of research has identified a group of genes that, when mutationally inactivated, cause hyper-mutation. As a result, these gene products were named “Mut” proteins and are the key players in mismatch repair function (Figure 4). Three proteins that are required for detecting and repairing damage are MutS, MutH, and MutL. During this process, multiple mismatch repair proteins, like MutS homologs, join together to form heterodimers, proteins composed of two polypeptide chains. In eukaryotes, heterodimers MSH2-MSH6 and MSH2-MSH3 facilitate MMR. If there is a mispaired base, the MutS heterodimers will recognize and bind the damaged DNA. This binding activates MutL to bind the MutS-DNA complex and subsequently activates MutH to bind hemimethylated sites along the daughter strand. MutH then nicks the DNA and recruits DNA Helicase II (UvrD helicase) to separate the strands. The newly formed MutSHL complex moves along the daughter DNA toward the mismatch and excises the mismatch and its surrounding nucleotides. DNA polymerase, DNA ligase, and Dam methylase then fill any single stranded gaps, seal nicks, and methylate the daughter strand, respectively.[6] If there is DNA damage, the heterodimers will again bind to the DNA and signal cell cycle arrest where the cell can either repair the damage or prompt apoptosis.[7] As a result of MSH2 deleterious mutations, initial mismatch recognition is mutated because the MutS heterodimer is mutationally inactivated. Thus, MutL and MutH cannot bind to the damaged DNA and rate of point mutations in genes substantially increases along with microsatellite instability and the probability that a cancerous mutation will form.


200902251425_55183_000.jpg
Figure 5. Depiction of Colorectal Stem Cells - from the colonic epithelium to the individual colonic crypt.
MSH2 is classified as a tumor suppressor gene or, more specifically, a caretaker gene. Caretaker genes are directly involved in repairing the DNA and, when mutated, can cause significantly higher mutation levels than other tumor suppressors like p53.[8] To avoid genetic aberrations like mismatched nucleotides and ensure DNA fidelity, human genomes have evolved to protect colonic epithelial stem cells in a variety of ways. Stem cells are “quiescent and rarely replicate” and hidden in tissues with low exposure to any environmental toxins or mutagens. Colonic epithelial stem cells, in particular, are located at the bottom of crypts, glands found in the epithelial lining of the small intestine and colon, and are protected by a thick mucin produced by neighboring cells (Figure 5). If they become damaged, they instigate apoptosis as opposed to repair.[9] However, these evolutionary defense mechanisms are rendered useless against hereditary MMR mutations. Since these mutations are already encoded in the genome, the ramifications from mutations like hereditary nonpolyposis colorectal cancer are not limited to cells frequently exposed to environmental toxins. This allows for the occurrence of MSH2 mutations in colonic epithelial stem cells and the evasion of any evolutionary measures taken to avoid damaged stem cells. Recent research has also proven that MSH2, in response to methylating agent damage, is required to induce both the “immediate and delayed responses of lower crypt [or colonic epithelial stem] cells.”[10] Thus, while the epithelial stem cells are already at great risk for genomic instability from a lack of DNA mismatch repair, their probability of tumorigenisis is further heightened as their ability to self-destruct is also blocked by MSH2 mutations. In a sense, the question of if a tumor will develop is less important than the question of when.

There is no current treatment for any hereditary nonpolyposis colorectal cancer, including MSH2 mutations. Most available chemotherapies target gain-of-function oncogenes where molecules downstream of the oncoprotein can be inhibited. However, treating tumor suppressor genes is indefinitely more difficult, especially in the case of deleterious mutations in tumor suppressor genes like MSH2. In deleterious mutations, the fundamental problem is that a functioning copy of the gene is missing from the chromosome. There is no hypermethylation or oncoprotein inhibiting the ability to produce viable MMR proteins and thus, these mutations cannot be treated with any type of drug. The only way to “fix” a MSH2 mutation is gene therapy, however, there is currently no such treatment for HNPCC. Instead of gene therapy, recent research has attempted to neutralize these “rogue genes” in an attempt to hit the “brakes” of cancer formation.[11] Much like an automobile, MSH2 mutations are gas pedals that cause acceleration towards cancer formation. Researchers hope to find a brake that, while not able to prevent cancer formation, will stall it so there is no tumor growth or recession. However, this technology is still years away and the genetic screening remains the closest “cure” to this disease.

Hereditary Nonpolyposis Colorectal Cancer



Jennie's deleterious mutation in MSH2 confered as much as an 82 percent risk of colorectal cancer and a 60 percent risk of endometrial cancer by age 70. Common signs for HNPCC are cancer diagnoses and deaths at younger than normal ages in multiple generations of a family. Although Lynch syndrome itself is not a disease, it is a genetic mutation that predisposes people to cancer, colorectal cancer in particular. Since it is an autosomal dominant genetic condition, it is likely that Jennie inherited the mutation from her mother, whose contraction of uterine and rectal cancer at such a young age are strong signs of HNPCC.[12] It is also likely that many of the patient’s siblings also inherited the genetic condition as the majority had colon cancer and died at an early age.
Lynch syndrome is relatively uncommon—in the United States alone, approximately 140,000 new cases of colorectal cancer are diagnosed yearly and only about 3-5 percent of these cases are caused by Lynch syndrome.[13] As HNPCC is a hereditary disorder, the only prevention methods are to test for DNA mutations through the genetic screening of children predisposed to Lynch syndrome. Although patients presenting with HNPCC are inevitably at higher risk to contract cancer, “a diet low in animal fats and high in fruits, vegetables, and whole grains” and a lifestyle of increased physical activity, of low alcohol consumption, and without smoking can reduce the risk of cancer. However, the most effective way to combat colon cancer is to get regular screening for precancerous polyps. Treatment is most effective if these polyps can be detected at an early stage.[14]
Jennie’s symptoms of low appetite, fatigue, bloody stool, low-grade microcytic anemia, abdominal cramping, and bowel pressure are consistent with reported signs of colon cancer. Blood in the stool is often caused by growths near the end of the left colon or rectum that prevent proper passage of bowl. These bowel obstructions can also cause constipation, cramping, and the feeling that the bowel is not completely empty, much like Jennie was experiencing. Because of a slow loss of blood that can occur over long periods of time, colon cancer can also cause iron deficiency that would explain the patient’s low-grade microcytic anemia and fatigue.[15] As Jennie’s colon cancer was stage IIIC, it has spread through the serosa of the colon wall and infected several lymph nodes, but has not yet spread to other organs (Figure 6).[16] When detected at a localized stage, the 5-year survival rate is 90 percent; however, 60 percent of colon cancer cases, like Jennie’s, are diagnosed after the cancer has spread to nearby organs or lymph nodes and the 5-year survival rate drops to 71 percent. Because of Jennie’s age and the stage of her cancer, her 5-year survival rate was likely to be even less.

CDR688436-750.jpg
Figure 6. Stage IIIC colorectal cancer.

Although those who suffer from Lynch system usually develop cancer at an early age, Jennie developed colon cancer much later than most people. This is not unusual as she did previously develop three other cancers prior to the colon cancer. Additionally, there is no specific age for cancer development in Lynch syndrome. People diagnosed with HNPCC usually get cancer at an earlier age because their MMRs cannot fix any mutations to the DNA and transcription goes unregulated. This presents cancer with the perfect opportunity to take seed and grow into a tumor. However, this just means people with Lynch are more likely to get cancer earlier. In Jennie’s case, her body was able to fend off colon cancer just a little longer than most people with HNPCC.


Treatment: Chemotherapy with 5-fluorouracil and a Colostomy



Screen Shot 2015-04-27 at 9.48.47 PM.png
Figure 7. September 30, 2012 colonoscopy results.



COLONOSCOPY RESULTS: Colonoscopy on September 30, 2012 showed sessile polyp with a mucosal abnormality, 2mm by 3mm, in the proximal ascending colon and two protruding, medium-size, malignant appearing, ulcerated masses with stigmata of bleeding, one in the rectum and the other in the anus (Figure 7).

Figure 8. Colostomy diorama.
Figure 8. Colostomy diorama.









Because of her age and the progression of her colorectal cancer, Dr. Smith could not offer Jennie the standard of care or the best of treatments. In a person twenty years younger, the standard of care would entail removing the whole rectum area and leaving a colostomy—a procedure where a stoma is created in the abdomen wall and the edges of the bowel are stitched to the skin. If the tumor was not so close to her anus, a surgeon could perform a surgical anastomosis or temporary colostomy—a procedure where the unhealthy section of the colon is removed and the two healthy ends are rejoined (Figure 8). However, in Jennie’s case, the standard of care would mean hours of anesthesia that could lead to heart attacks, strokes, and other complications that could arise during surgery. At the other extreme, Jennie could opt out of receiving treatment, but doing so would only give her about another year to live .[17]

Instead, Dr. Smith decided to take a pathway between these two extremes—Jennie would receive radiation coupled with chemotherapy. With this particular treatment, there would be a 20 percent chance of putting Jennie’s cancer in remission or getting rid of it completely. As nausea would be a likely side effect of the chemotherapy, Dr. Smith also prescribed ondansetron and prochlorperazine, drugs that work as receptor antagonists to prevent vomiting.[18]

Dr. Smith prescribed an oral form of chemotherapy, Xeloda, for Jennie to take on the days she went in for radiation. Xeloda, the commercial name for a drug called capecitabine, is converted to 5-fluorouracil (5-FU) upon absorption through the intestinal mucosa. 5-FU is a thymidine synthase inhibitor (TSI) and its presence in a cell inhibits “the only means of adding a methyl group to the 5-position of the pyrimidine ring in the de novo synthesis of thymidine.”[19] In short, 5-FU prevents the formation of the T nucleotide. As a result, DNA synthesis cannot proceed and any fast growing tumor cells are forced to stop growing. 5-FU can be administered in a variety of ways ranging from intravenously and intramuscularly to orally. Jennie received the orally administered capecitabine specifically because she had colorectal cancer. Since capecitabine is not metabolized until absorption into the intestine, it is particularly advantageous in treating colorectal cancer for two reasons: first, there would be an increased drug concentration at the cancer site and therefore greater anti-tumor activity, and second, there would be reduced drug levels in non-cancerous tissue. Since 5-FU targets all cells, this localized chemotherapy would cause Jennie’s body the least damage.[20] Preclinical data has suggested an improved efficacy for capecitabine over fluorouracil as xenograft models show 5-FU concentrations higher in the tumors than in other healthy tissue.

Although capecitabine has suggested improved efficacy in treating colon cancer, Jennie’s cancer did not show significant improvement. Although there is no definitive proof, it is likely that Jennie’s cancer did not respond to the capecitabine because of her hereditary non-polyposis colorectal cancer as recent studies have produced conflicting evidence on the effects of 5-FU on HNPCC tumors. HNPCC is characterized by high microsatellite instability (MSI-H) as a result of deleterious mutations in mismatch repair proteins (MMR deficiency). Since mismatch repair proteins are responsible for recognizing and correcting damaged DNA, having MMR deficiency leads to an accumulation of mutations that increases the likelihood of contracting a cancerous mutation. Although not all tumors in HNPCC patients exhibit MSI-H, it is likely that at least one of Jennie’s tumors is MMR deficient. Although not required, “most MSI-H tumors show a characteristic phenotype including right-sided location […] and mucinous differentiation,” two characteristics observed in Jennie’s colonoscopy (Figure 7).[21] Some evidence suggests that patients, particularly those under fifty years of age, with MMR deficient tumors have a more favorable prognosis for survival. However, other studies have produced evidence that patients enrolled in randomized treatments of 5-FU had survival benefit for those with MMR proficient tumors, but not MMR deficient tumors.[22] Although 5-FU’s effect on MSI-H tumors is unclear, it should be noted that this study included MMR deficient tumors of both sporadic origin and HNPCC origin—a fact that may have produced conflicting results.

Screen Shot 2015-04-23 at 1.48.13 PM.png
Figure 9. Carethers (2006).
A recent study out of Amsterdam that focused exclusively on tumors of HNPCC origin found that 5-fluorouracil is ineffective in improving 5-year survival rates in patients with MSI-H tumors of HNPCC origins. When tracking survival rates in these tumors, both cumulative survival and disease-free survival rates for those who received 5-FU were not significantly different from those who did not (Figure 9). However, the retrospective nature of this study “demands confirmation of the results by further studies that should elucidate the role of MSI in 5-FU treatment of MSI-H tumors in HNPCC.”[23] If a functioning MMR system is required for 5-fluorouracil cytotoxicity, then 5-FU is not likely to be beneficial to HNPCC patients. However, other clinical studies have produced evidence that contradict this study’s findings so all possible third variables that could affect 5-FU cytotoxicity must be evaluated.
Some evidence suggests that MMR proteins recognize 5-FU and incorporate it into DNA and thus with MMR deficiency, 5-FU goes unbound with “no subsequent triggering for tumor cell demise.”[24] Although there is no definitive evidence that proves this hypothesis, it is likely that Jennie’s HNPCC condition prevented the 5-fluorouracil from stopping cell growth in a double negative feedback loop. As a result, Dr. Smith concluded that Jennie’s only remaining option was surgery. However, Dr. Smith, Jennie, and her family were all extremely cautious about exploring this area of treatment especially because of her age. The recovery process would be grueling and they had to weigh these painful months against letting the cancer take its course. But like everything else in her life, Jennie decided to go forward with the surgery. On March 24, 2014, Jennie underwent a colostomy to remove the tumor, however, because of its location near the anus and subsequent inability to perform surgical anastomosis, she received an ostomy pouching system—a bag placed around the stoma to collect waste from the colon—for waste disposal. The months following the surgery were as painful as was expected, but Jennie never regretted her decision to fight. After about three months, Jennie began to regain the weight she lost and was able to walk on her own. Despite occasional infections from the colostomy, Jennie is currently free of her colorectal cancer and living comfortably.

Afterward



Not all people are as lucky as Jennie—every year, about 50,000 people die from colorectal cancer in the United States. However, cancer research has allowed doctors to more effectively treat their patients by giving people an understanding of what is happening at the molecular level. Despite the millions of lives it claims each year, cancer is a complex ecosystem that finds a way to survive despite human efforts to cut it down. In order to fight this disease, doctors and researchers must first come to understand how it operates and why it does what it does. They have been able to develop treatments like chemotherapy and radiation that allow doctors to kill cancer, however, not without also hurting the patient. Although there has been numerous cancer breakthroughs in the past decades, we are not much closer to finding a "cure" than when we started. If we have learned anything, however, it is that humans are as determined as cancer, if not more, to survive. While we cannot win every battle, we must keep fighting.
"Given her age, the severity of her colorectal cancer, and the presence of bladder cancer, non-Hodgkin’s lymphoma, and skin cancer, Jennie is truly a miracle patient—a description that can be attributed to her resilience and will to survive. Throughout all my years of practice, I have never met someone who was so dedicated to her family, her health, and her life. If I ever saw someone beat cancer, it was Jennie." Dr. Smith



  1. ^ American Cancer Society. Cancer Facts & Figures 2015. Atlanta: American Cancer Society; 2015. Retrieved May 31, 2015, from
    http://www.cancer.org/acs/groups/content/@research/documents/document/acspc-042801.pdf
  2. ^ Cancer Facts. (n.d.). Retrieved June 1, 2015, from http://www.thomlatimercares.org/Cancer_Facts.htm
  3. ^ World Cancer Day 2013 - Global Press Release. (2013). Retrieved June 1, 2015, from http://www.worldcancerday.org/world-cancer-day-2013-global-press-release
  4. ^ Hanahan, D., & Weinberg, R. (2011). Hallmarks Of Cancer: The Next Generation. Cell, 646-674.
  5. ^ Gulati, S., Gustafson, S., & Daw, H. (2011, September 1). Lynch Syndrome Associated With PMS2 Mutation: Understanding Current Concepts. Retrieved April 17, 2015, from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3269140/
  6. ^ Kunkel, T., & Erie, D. (2005). Dna Mismatch Repair*. Annual Review of Biochemistry, 74, 681-710. doi:DOI: 10.1146/annurev.biochem.74.082803.133243. Retrieved May 20, 2015, from
    http://www.annualreviews.org/doi/full/10.1146/annurev.biochem.74.082803.133243?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%3dpubmed&
  7. ^ Hargreaves, V. V., Shell, S. S., Mazur, D. J., Hess, M. T., & Kolodner, R. D. (2010). Interaction between the Msh2 and Msh6 Nucleotide-binding Sites in the Saccharomyces cerevisiae Msh2-Msh6 Complex. The Journal of Biological Chemistry, 285(12), 9301–9310. doi:10.1074/jbc.M109.096388. Retrieved May 20, 2015, from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2838348/
  8. ^ Deininger, P. (1999). Genetic Instability in Cancer: Caretaker and Gatekeeper Genes. The Ochsner Journal, 1(4), 206–209. Retrieved May 20, 2015, from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3145442/
  9. ^ Armaghany, T., Wilson, J. D., Chu, Q., & Mills, G. (2012). Genetic Alterations in Colorectal Cancer. Gastrointestinal Cancer Research: GCR, 5(1), 19–27. Retrieved May 20, 2015, from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3348713/
  10. ^ Toft, N. J., Winton, D. J., Kelly, J., Howard, L. A., Dekker, M., te Riele, H., … Clarke, A. R. (1999). Msh2 status modulates both apoptosis and mutation frequency in the murine small intestine. Proceedings of the National Academy of Sciences of the United States of America, 96(7), 3911–3915. Retrieved May 18, 2015, from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC22394/
  11. ^ Genetic Testing. (2013, May 24). Retrieved May 25, 2015, from http://www.lynchcancers.com/index.php/genetic-testing
  12. ^ Lynch syndrome. (2015, April 6). Retrieved April 9, 2015, from http://ghr.nlm.nih.gov/condition/lynch-syndrome
  13. ^ Ibid.
  14. ^ What Can I Do to Reduce My Risk of Colorectal Cancer? (2014, April 2). Retrieved April 9, 2015, from http://www.cdc.gov/cancer/colorectal/basic_info/prevention.htm
  15. ^ Signs and Symptoms of Colon Cancer. (n.d.). Retrieved April 9, 2015, from http://www.cancer.org/cancer/news/signs-and-symptoms-of-colon-cancer
  16. ^ Colon Cancer Treatment. (2015, April 2). Retrieved April 10, 2015, from http://www.cancer.gov/cancertopics/pdq/treatment/colon/Patient/page2
  17. ^ Sumida, K. (2013, November 25). [Personal meeting].
  18. ^ Ibid.
  19. ^ Papamichael D (2000). "The Use of Thymidylate Synthase Inhibitors in the Treatment of Advanced Colorectal Cancer: Current Status". NCBI 4 (6): 478–87. Retrieved April 22, 2015, from http://theoncologist.alphamedpress.org/content/4/6/478.long
  20. ^ Ibid.
  21. ^ Turner, J., Baehner, F., Bloom, K., Caughron, S., Cooper, T., Friedberg, R., . . . Jenkins, C. (2011). Prognostic Uses of MSI Testing. College of American Pathologists, 1. Retrieved April 22, 2015, from http://www.cap.org/apps/docs/committees/technology/microsatellite_testing.pdf
  22. ^ Carethers, J. M. (2006). Prospective evaluation of fluorouracil chemotherapy based on the genetic makeup of colorectal cancer. Gut, 55(6), 759–761. doi:10.1136/gut.2005.085274.Retrieved April 22, 2015, from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1856230/
  23. ^ 2004, January 4). Survival after adjuvant 5-FU treatment for stage III colon cancer in hereditary nonpolyposis colorectal cancer. Retrieved April 23, 2015, from http://onlinelibrary.wiley.com/doi/10.1002/ijc.11712/abstract;jsessionid=32597BFB0C0074CE72C38F2A93DB57EB.f02t03
  24. ^ Carethers, 2006.