A History of Hereditary Colorectal Cancer

Introduction

Hereditary colorectal cancer is important for two main reasons: firstly, the impact that it has on patients and families affected by its syndromes, and secondly, the lessons it teaches about the genetic origins of colorectal cancer. The stories of how hereditary syndromes of colorectal cancer were discovered, researched and genetically defined feature serendipity, innovative thinking and hard work. This chapter reviews these stories and highlights some of the personalities involved. Hereditary colorectal cancer was not discovered in a vacuum however, and the story of the syndromes is presented within a broader view of developing medical science. Table 1 presents a timeline of events.

Human Genetics

Human inheritance is obvious in the way children resemble their parents, and yet for millennia nobody understood how that happened. Finally, in latter half of the 19th century, the penny started to drop. An Austrian Monk named Gregor Mendel studied peas in his Monastery’s garden and noted how crossbreeding affected various traits. Using these observations, he defined the patterns of inheritance coining the terms “dominant” and “recessive” inheritance. He realized that “invisible factors” (i.e. genes) were the mechanism of the inheritance. When Mendel became Abbott in 1868, he had to stop his scientific work and his ideas were neither universally accepted nor promoted.

At about the same time as Mendel became Abbott a Swiss biologist named Friedrich Mieschler was working in Felix Hoppe-Seyler's laboratory at the University of Tubingen, studying the nuclei of white blood cells. He obtained the cells from pus that he washed off bandages at the nearby hospital. He found a phosphate rich chemical that he called nuclein. This turned out to be nucleic acids, the substance of DNA. Later on, he suggested that “nuclein” was involved in heredity.

By now, people were thinking about heredity and how it worked. It wasn’t long before the Dutch botanist de Vries called the “invisible factors” of Mendel “pangenes” and in 1905 the Danish geneticist Wilhelm Johannsen coined the term “gene”. The role of chromosomes in heredity was appreciated early in the 20th century and the word “genetics” was first used by the English biologist William Bateson in 1906. Appreciation of the location of genes within DNA, of the structure of DNA by Watson and Crick, and the mechanism of protein manufacture via gene transcription and RNA translation occurred during the 1940’s and 50’s, and led to a level of understanding that allowed molecular biology to take off in a series of spectacular advances. Highlights included the discovery of frameshift mutations in 1961, the function of messenger RNA in the same year, and a definition of the genetic code by 1967. Genetic sequencing began in 1972 and in 1977 Fred Sanger, Walter Gilbert and Allan Maxam sequenced DNA. The era of Genomics had arrived.

In 1983 the polymerase chain reaction was invented by the American biochemist Kary Banks Mullis, laying the foundation for clinical genetic testing. 1987 saw the introduction of the first automated DNA sequencing machine (Applied Biosystems) and in 1990 the Human Genome Project began and the concept of introns and exons was described. The 1990’s were a truly significant decade for the genetics of hereditary colorectal cancer. In 1991, APC was cloned, to be followed by the localization and cloning of the mismatch repair genes that cause Lynch syndrome.

DNA Methylation

Addition of a methyl group to DNA (at CpG dinucleotides) was first found to influence gene expression in 1975. Research into cancer-related DNA methylation began in 1983 when Feinberg and Vogelstein described hypomethylation in colorectal cancer cells, but they did not know if this was a cause or an effect of the cancer. Focal hypermethylation in cancers was found by Stephen Baylin in 1986. This significance of this finding became apparent in 1989 when promoter methylation of the recently cloned RB1 was linked to loss of expression in some retinoblastomas. This kicked off the acceptance of the role of hypermethylation in carcinogenesis. Measurement of methylation in tumors was made practical in 1992 when Frommer used the sodium bisulphite reaction as a way of examining methylation patterns in genomic DNA. This led to studies such as those detailing age-related changes in colorectal DNA methylation and promoter methylation of MLH1 in some colorectal cancers.

Genetic Testing

One of the earliest and simplest genetic tests was a karyotype, where chromosomes were separated and stained. While karyotype analysis dates from 1922 it was not until 1956 that the true number of human chromosomes was defined. Then structural abnormalities such as duplication, transversion and deletion were obvious and could be related to disease. Linkage analysis was another early, non-sequencing test, which relied on the observation that genes close together on a chromosome tended to be inherited together during cell division. This concept, developed by Thomas Hunt Morgan, led to the LOD score (logarithm of odds) used by Newton Morgan to determine the strength of linkage. Protein truncation testing was popular in patients with FAP before the causative gene was cloned, because most variants in APC cause “stop” codons that prematurely ended protein production. However after 1991, with cloning of APC and the mismatch repair genes of Lynch syndrome, gene sequencing began to be increasingly popular. In 2000, massively parallel signature sequencing was described and became the first form of “next generation sequencing” (NGS) technology. Three years later Francis Collins presented for the first time the DNA sequence of the human genome. It was not until 2012 that NGS became established enough for the era of genetic panel testing to begin clinically, leading to ever decreasing costs and increasing availability of genetic testing.

Genetic Counseling

At the turn of the 20th century, William Bateson suggested that the study of heredity be called “genetics”. At the same time the appreciation of inheritance as a way of transmitting familial traits from parent to offspring led to the start of offering advice to people based on the traits in question. This was the start of Genetic Counseling. Initially the medical aspects of inheritance became subservient to social considerations, resulting in the field of eugenics. This led to forced sterilization and ultimately euthanasia of people deemed “defective.” As clinical studies of heredity and genetics progressed, medical aspects of genetic counseling became more prominent. Sheldon Clark Reed used the term “genetic counseling” in 1947 and produced a book called Counseling in Medical Genetics 8 years later. A Master’s degree in Genetic counseling was offered at Sarah Lawrence College in Bronxville, New York in 1969. Ten years later the National Society of Genetic Counselors was founded.

In vitro fertilization and preimplantation genetic diagnosis

In-vitro fertilization is sometimes used in patients with FAP to prevent implantation of an embryo affected with the disease. The first IVF pregnancy happened in Australia in 1973 and the first successful pregnancy in 1978. PGD has been used in patients with FAP since 1998.

Gene Therapy

The first successful introduction of human genes into the nucleus of cells occurred in 1989 and the first therapeutic use of the procedure happened a year later. The first use of gene therapy for cancer in an inherited syndrome was in 2008, in a patient with Li Fraumeni syndrome using an adenovirus to carry a replacement for mutant TP53.

Clinical Discoveries and Hereditary Colorectal Cancer

The phenomenon of intestinal polyps occurring in multiple young people within the same family was first described in the late 19th century. These isolated reports set the stage for a growing realization that precancerous and cancerous lesions can be inherited. For FAP this was codified in 1927 by the British physician E.A Cockayne, who ascribed Mendelian dominant inheritance to the condition. Meanwhile, on the other side of the Atlantic Ocean, an American pathologist had been researching a large German family with multiple relatives affected by colon, uterus and stomach cancer. That pathologist was Aldred Warthin. He suggested that cancers could be familial and that this was associated with a young age at diagnosis. He was largely ignored until the 1960’s. By contrast, the clinically obvious inheritance of adenomatous polyposis became the stimulus for the formation of the first registry for hereditary colorectal cancer at St. Mark’s Hospital in London in 1924. Here, under the guidance and leadership of surgeon J.P. Lockhart Mummery, pathologist Cuthbert Dukes and “lab boy” H.J.R. Bussey, began what would become the leading center for familial adenomatous polyposis research and management in the world. The concept of documenting family pedigrees and using them to identify relatives at risk was key to managing the disease and facilitated programs of surveillance and prophylactic surgery. Bussey’s carefully recorded database led to multiple helpful studies and the St. Mark’s registry became the model for registries in Cleveland, New York City, Baltimore, Boston, Houston, Toronto, and elsewhere. St Mark’s trainees became leaders in their own right, many of whom would use their experience to establish hereditary colorectal cancer registries where they settled.

The first surgery recorded for patients with FAP was a colectomy, performed at St. Mark’s hospital in 1918. The first total colectomy with ileorectal anastomosis was reported in 1948 by O.V. Lloyd Davies from St. Mark’s Hospital, and in 1952 the Brooke ileostomy was described by English surgeon Brian Brooke. The specialty of enterostomal therapy was started by Rupert Turnbull at the Cleveland Clinic at about the same time. A completely new option for patients needing proctocolectomy was introduced in 1969 by Nils Kock in Gothenburg, Sweden who described his continent ileostomy. Ten years later, FAP surgery entered the modern age when John Nicholls and Sir Alan Parks from St. Mark’s Hospital first performed the ileal pouch anal anastomosis. In 1980 Joji Utsinomiya from Kobe in Japan invented the J pouch, still the favorite way of dealing with loss of the entire large intestine. Soon after the introduction of the pouch came surgical staplers, brought to the US from Russia by Mark Ravitch in 1979.

In 1975, W.R.Waddell, a surgeon at the University of Colorado, used cyclic 3’,5’ AMP inhibitors to treat a patient with abdominal desmoid disease. It worked. In 1980 he treated three patients with indomethacin and ascorbate, with responses in all three. In 1983 Waddell used sulindac successfully to suppress polyposis and introduced the use of tamoxifen to treat desmoids. Both sulindac and tamoxifen are used in FAP patients to this day.

The next major development in surgical technique came in 1991 when Moises Jacobs in Miami reported a series of laparoscopic colectomies. This minimally invasive approach was soon applied to patients with FAP at the Cleveland Clinic, with Jeff Milsom reporting a series of ileorectal anastomoses in 1997, and Peter Marcello reporting laparoscopic assisted pouch surgery three years later.

Upper GI surgery in FAP came of age in 1935 when Alan Whipple in Columbia University, New York, described a practicable way of removing the duodenum and pancreas. The pancreas-preserving duodenectomy came 60 years later, thanks to Ray Chung at the Cleveland Clinic, and has been popularized by Matthew Walsh at the same institution.

Endoscopy

The use of scopes to see into the GI tract has always been limited by the lack of an ability to transmit useful light more than a few centimeters. Rigid, poorly illuminated scopes were not comfortable to use and did not provide a decent view. Progress was finally made in 1932 when Wolf and Schindler developed a semiflexible gastroscope, however the world of gastroenterology was waiting for Basil Hirschowitz, who in 1958 produced his flexible fiberoptic gastroscope. Hirschowitz was a South Africa from Bethal, a small farming town east of Johannesburg. He moved to USA in 1953 at the age of 28. His use of fiberoptics was soon applied by Japanese scientists to produce a flexible endoscope, and this device was used in the colon by Japanese surgeon Hiromi Shinya who was working at Beth Israel Hospital in Boston. He joined with American Surgeon William Wolff to pioneer endoscopic polypectomy, reporting this in 1969. The next major advance was the use of a CCD in endoscopes, replacing the fiberoptics and allowing greater freedom in scope design and electronic manipulation of the image. Initial use was described by Mike Sivak in Cleveland in 1984. Accessory techniques such as chromoendoscopy, endoscopic mucosal resection (1988), and endoscopic submucosal dissection (2003) have arrived and are in general use. In the upper tract endoscopic ampullectomy and was described in 1983 by Suzuki, and the Spigelman Criteria for measuring duodenal adenomatosis came in 1998, the product of a paper from St. Mark’s hospital the lead author of which was an Australian fellow, Alan Spigelman.

Familial adenomatous polyposis

Familial adenomatous polyposis (FAP) was the first hereditary colorectal cancer syndrome to come to the attention of clinicians. We have followed its progress from the first surgery at St. Mark’s in 1918, to the first registry in 1925. The importance of extracolonic manifestations in the diagnosis and management of FAP was noted in 1950 by Eldon Gardner. He reported a large family affected by intestinal polyposis with apparent dominant inheritance of the disease. This was not news. What was news was the osteomas, fibromas and epidermoid cysts that were inherited with the polyposis. In 1958 W.G. Smith named this syndrome after Gardner, an eponym that can be confusing to patients. The fibromas originally described by Gardner are in fact desmoid tumors. Desmoid tumors were initially described in 1832 and the name “desmoid” first used in this context in 1838. James Church at Cleveland Clinic developed a desmoid staging system to help classify patients with desmoid tumors and to guide their treatment as well as assign prognosis in 2005. Desmoids remain a serious challenge in patients with FAP. 1959 saw the eponym Turcot’s syndrome used to define gastrointestinal polyposis and brain tumors, although Crail had described the association 10 years before the Canadian, Turcot. The search for a gene responsible for FAP was guided in 1987 by Lemuel Herrera’s report of polyposis in a patient with an interstitial deletion of the long arm of chromosome 5. Linkage analysis directed attention to the 5q21-22 region and in 1991 the gene was located and cloned. Soon after its discovery Nagase reported on genotype-phenotype correlation with polyp count and since then other correlations have been found with Gardner’s Syndrome, congenital hypertrophy of the retinal pigmented epithelium and desmoid severity. In 2011 the Swedish group led by Nordling showed that deletions in promoter 1B, one of two promoters of APC, cause polyposis. This was important as it explained some families with FAP but no detectable variant in the gene.

Peutz-Jeghers Syndrome

In 1921 Dutch internist Jan Peutz published a case report describing the syndrome later named for him. Twenty-eight years later, Harold Joseph Jeghers, along with his co-authors Kermit Katz and Victor McKusick, described the syndrome in more detail, and in 1998 variants in STK11 were associated with inheritance of the syndrome. Initially patients presenting with small bowel obstruction from intussusception of a polyp just had the symptomatic polyp resected, but in 2004 the advantage of the “clean sweep” method of removing all polyps instead of just the large or symptomatic ones was published by surgeons from the Cleveland Clinic.

Juvenile Polyposis

Juvenile polyps were first described by Diamond in 1939, although he called the lesions “adenomas.” They were first recognized as a distinct entity by Horilleno 18 years later, and juvenile polyposis was reported in 1964. Its autosomal dominant inheritance was recognized in 1966, but it was not until 1998 that James Howe associated variants in SMAD4 with the syndrome. Three years later BMPR1A was added as a causative gene.

Cowden Syndrome

Cowden Syndrome is a condition of multiple hamartomas and neoplasms of the skin, thyroid, breast, and colon, in addition to craniofacial abnormities. It is named after the index patient, Rachel Cowden, who presented in Youngstown, Ohio to Dr. Kenneth Lloyd, a dermatologist. He reported this case with Dr Mecey Dennis, in 1962. Ten years later the syndrome was classified as an autosomal dominantly inherited hamartomatous growth disorder. Charis Eng, from the Cleveland Clinic, was the first to identify variants in PTEN as the cause of Cowden Syndrome. She also reported its role in Bannayan Ruvalcaba Riley Syndrome, a related condition of hamartomatous polyposis. Eng led the effort to define the clinical criteria for these hamartomatous syndromes. In 2007 they were combined as the PTEN Hamartoma Tumor Syndrome (PHTS).

More recent hereditary polyposes

In the last decade there have been at least five newly discovered hereditary syndromes of intestinal polyposis. They are generally syndromes of oligopolyposis and colorectal cancer, they are uncommon, and three of them are recessively inherited. The first was MUTYH associated polyposis, described in 2003 by the Cardiff group led by Julian Sampson. This attenuated form of FAP was caused by a recessively inherited variant in MUTYH, a gene involved in base excision repair. NTHL1, another gene in the base excision repair pathway was found also to be involved in autosomal recessive inheritance of oligopolyposis and colorectal cancer, as reported by a Dutch group in 2015. In 2012, hereditary mixed polyposis syndrome, where dominantly inherited polyposis features a mixture of adenomas, hamartomas and serrated polyps, was linked to a duplication in SCG5, upstream of the GREM1, that results in an increased expression of GREM1. This is predicted to stimulate bone morphogenic protein pathway activity (also increased in juvenile polyposis). Also in 2012, Ian Tomlinson’s group reported a new syndrome of early age of onset colorectal and endometrial cancers, and advanced adenomas due to a germline pathogenic variant in two DNA polymerase genes, POLD1 and POLE. The most recent syndrome, reported in 2016, is due to recessive inheritance of pathogenic variants in MSH3, a mismatch repair gene correcting mistakes in large microsatellites. Tumors in patients with MSH3 polyposis show Elevated Microsatellite Alterations at Selected Tetranucleotide repeats (EMAST).

Lynch Syndrome

The story of Lynch syndrome began with Aldred Warthin in 1913, but largely died with him in 1931. However, it was picked up again 30 years later by Henry Lynch, a medical oncologist. At about that time Lynch had noted the familial occurrence of colon cancers and started to research the possibility that cancer could be hereditary. He found Warthin’s family G and persisted with his research in the face of cynicism and disbelief. In 1971 he published “Family G Revisited”, using the term “cancer family syndrome” to refer to the phenomenon. In 1973 Rick Boland described two additional families and used Lynch syndrome I and II to distinguish families with a colorectal cancer only phenotype from those with cancers in other organs. In 1985 Lynch changed the name of the putative syndrome to Hereditary Non-Polyposis Colorectal Cancer (HNPCC) to distinguish it from FAP. HNPCC lacked specific diagnostic criteria until 1991 when Hans Vasen reported on the results of a meeting where the Amsterdam Criteria were defined. These criteria were intended for research but quickly became used clinically to define “Lynch” families. There was a lot of interest in HNPCC, and discoveries started happening quickly. The phenomenon of microsatellite instability (MSI) was noted in a subset of colorectal cancers by Manuel Perucho and Steven Thibodeau, who suggested a novel form of colorectal carcinogenesis. Coincidentally a group from Finland linked MSI to hereditary colorectal cancer. Within 2 years MSH2, MLH1 and PMS2 had been identified as the causes of what would become known as Lynch syndrome. Identification of MSH6 came in 1997 and EPCAM in 2009. In 1997 there was a meeting in Bethesda to decide on criteria for performing MSI analysis as a screening test for tumors, resulting in the Bethesda guidelines. These were revised in 2004 while the original Amsterdam criteria had been revised in 1999. Subsequent progress in genetic testing has diminished the value of these clinical guidelines.

Soon after discovery of the “Lynch” genes, antibodies to the Lynch proteins were produced and used in immunohistochemical (IHC) studies of tissues. In 2005 Hampel reported the results of screening tumors with IHC and MSI, showing that patients with Lynch syndrome often did not meet Amsterdam or Bethesda criteria. This showed the value of such screening, which has been adopted broadly as a way of diagnosis. In 2005 Laney Lindor reported on a group of families that met Amsterdam criteria but had microsatellite stable cancers. These families did not have Lynch Syndrome, which now came to be defined by a germline variant in a mismatch repair gene. Lindor called them “Familial Colorectal Cancer Type X”.

Subsequent progress in Lynch syndrome has been mainly clinical and epidemiological, using large databases to refine cancer risk according to genotype, and determining the best way to control this risk. Chemotherapy with 5FU was shown to be ineffective in Lynch patients, whose tumors had a better than average prognosis when compared to MSS tumors. The immunogenic effect of frame shift polypeptides produced by MSI cancer cells may explain this. In 2015, work on PD-1 blockage in patients with MSI cancers confirmed these observations. In 2011 the British geneticist Sir John Burn found a beneficial effect of prophylactic aspirin on the incidence of all Lynch cancers over long term follow-up (CAPP 2 study). A dose finding study is in progress.

Registries and Collaborative Groups

Lockhart-Mummery and his colleagues at St. Mark’s Hospital set the precedent for the registration and organized screening of patients at risk, and so showed the benefits of a Polyposis Registry. This concept spread through visitors and trainees from St. Mark’s such as David Jagelman (Cleveland), Finlay Macrae (Melbourne) and Allan Spigelman (Newcastle). The Johns Hopkins Registry, developed under Victor McKusick and Anne Krush and led by Francis Giardiello, was an early and productive center while other registries were started in Sweden, Holland, Singapore, South America, Toronto, Houston and Boston. As groups of dedicated caregivers and scientists dotted the globe, there was an urge to come together to talk about the conditions they served. The Leeds Castle Polyposis Group was started in 1985 by Ian Todd, senior surgeon at St. Mark’s, who wanted to discuss the thorny problem of desmoid disease in FAP. The International Collaborative Group on HNPCC was founded at a meeting in Israel in 1989 to provide a forum for discussion and a resource for collaborative studies in HNPCC. These two groups came together in 2003 as InSiGHT, the Internal Society for Gastrointestinal Hereditary Tumors. Regional societies were developed in 1995 for the Americas (Collaborative Group of the Americas on Inherited Gastrointestinal Cancer, CGA-IGC), 2006 for Europe (Mallorca Group and then European Hereditary Tumor Group, EHTG) and South America (Hereditary Tumors Study Group, GETH). The NCI-funded multicenter and multinational Collaborative Family Registry was particularly prolific and developed a large resource in data and specimens of patients with many of the syndromes.

The Future

Since the first experiments on peas performed by Mendel, progress in genetics and genetic diseases has increased exponentially. We are now on the steep slope of that curve, with advances in genetic technology and an increasing knowledge base maintaining the momentum. The future is exciting but it remains built on the foundations of the past.

Aldred Warthin