Of every 100 people with colon cancer, about 15 will have cancers that arise when mistakes in DNA during cell division are not caught and fixed.  Scientists call this defective mismatch repair or dMMR.

More often, colon cancer occurs when mutations in chromosomes accumulate but DNA repair pathways remain intact and mismatch repair is proficient (pMMR). This is true for about 85 percent of colon cancer.

Both prognosis and the potential benefit from FU-based chemotherapy appear to be very different for these two types of colon cancer. Knowing mismatch repair status of colon tumors can help patients and their doctors make better treatment decisions.

Patients with defective mismatch repair have better disease-free and overall survival and don’t seem to benefit from 5-FU at either stage II or stage III.  Stage II patients with dMMR have significantly poorer overall survival if they get chemo after surgery.

Caution:  These results come from studies of 5-FU plus levamisole or 5-FU plus leucovorin.  They don’t include any information from the current standard treatments of FOLFOX or FLOX which contain oxaliplatin in addition to 5-FU and leucovorin.

Defective mismatch repair is uncovered when either tumors have microsatellite instability (MSI) or immunohistochemical tests can’t find the proteins that are expressed by genes that control mismatch repair — MLH1, MSH2, MSH6, and PMS2.  Measuring either MSI or lack of MLH1 or MSH2 expression gives similar results in deciding whether a tumor is mismatch defective or proficient.

Almost all defective mismatch repair tumors are located in the right side of the colon.  They are poorly differentiated, often have lots of mucus, and tend to be infiltrated with immune-system cells (lymphocytes). Some scientists speculate that it is this improved immune-response that gives them their survival advantage.

Daniel Sargent, Ph.D., and his team analyzed mismatch repair status in the tumors of 457 patients in five different clinical trials with stage II or III colon cancer who had received either FU-based chemotherapy after surgery to remove their cancer or surgery alone.  They then pooled their information with a group of 570 patients who had been analyzed earlier.  The entire set included 1,027 patients, including 165 (16 percent) with defective mismatch repair.

They analyzed the impact of either proficient or defective mismatch repair as both a prognosis marker (its effect on survival despite treatment) and predictive marker (if chemotherapy treatment changes outcomes).

Pooled Results

Mismatch Repair Status as a Prognostic Marker

• Patients with defective mismatch repair (dMMR) who didn’t get 5-FU had significantly better disease-free and overall survival than patients with proficient mismatch repair (pMMR).
• When patients got chemotherapy, mismatch repair status had no impact on survival.

Mismatch Repair Status as a Predictive Marker for FU- based chemotherapy

• There was no benefit of FU-based chemotherapy for either stage II or stage III colon cancer patients with defective mismatch repair.
• There was no benefit of FU chemo for stage II patients with proficient mismatch repair.
• Stage III patients with pMMR did benefit from chemotherapy with 5-FU.
• Stage II patients with dMMR had worse overall survival when they got 5-FU than when they had surgery alone.

Graph of Predictive Value of dMMR in Adjuvant Colon Cancer

A — Stage II, dMMR
B–  StageIII dMMR
C — Stage II pMMR
D — Stage III pMMR

Click on graph to enlarge it.

Dr. Sargent’s analysis confirmed an earlier study reported in 2003 in the New England Journal of Medicine by Christine Ribic that found no benefit to treatment with 5-FU for patients with stage II or III colon cancer with high microsatellite stability, who may possibly be harmed by  5-FU-based chemotherapy and have a better overall prognosis after diagnosis.

His team concluded,

In conclusion, this prospectively specified analysis of data from randomized, clinical trials provides independent, supportive evidence of the following: dMMR colon cancers have a favorable stage adjusted prognosis compared with the majority of colon cancers; and patients with dMMR colon cancers do not benefit from FU based adjuvant therapy.

Further, they point out,

These findings support the conclusion that average-risk patients with colon cancer who are considered for FU based adjuvant therapy should have the tumor MMR status assessed to inform the likelihood of patient benefit of chemotherapy. Our conclusions are restricted to patients being considered for single agent, fluoropyrimidine-based therapy (ie, patients with stage II disease), and the conclusions provide guidance as to who should not be treated (ie, the dMMR subset). We believe that dMMR status in the setting of stage II disease should be considered a clinically useful marker of tumor biology and represents an additional step in individualized cancer therapy.

SOURCE: Sargent et al.,  Defective Mismatch Repair As a Predictive Marker for Lack of Efficacy of Fluorouracil-Based Adjuvant Therapy in Colon Cancer. Journal of Clinical Oncology, Volume 28, Number 20, pages 3219-3226,July 10, 2010.

Image:  Figure #2, Sargent et al., JCO

Cancers that are linked to microsatellite instability (MSI) don’t appear to be influenced by obesity, strengthening the belief that MSI cancers come about differently than the average colorectal cancer.

Overall, in a recent study, body mass index and weight gain during adult life increased risk of colorectal cancer by about 30 percent for men and 20 percent for women. However, increased risk was limited to microsatellite stable or microsatellite low tumors.

Led by Peter T. Campbell, PhD, from the Epidemiology Research Program at American Cancer Society in Atlanta, a research team compared height and weight at age 20 and recent height and weight for nearly 1,800 people with colorectal cancer to figures for 2,700 of their sex-matched brothers and sisters.

They also analyzed tumors for microsatellite instability in 7 out of 10 of the group with cancer.

They discovered a 38 percent increased risk for cancer in patients with microsatellite stable tumors, a 33 percent increased risk in those who had a low number of microsatellite markers, and no increase those whose tumors had high number of microsatellite changes (MSI).

Campbell and his colleagues concluded,

The increased risk of colorectal cancer associated with a high BMI might be largely restricted to tumors that display the more common MS-stable phenotype, suggesting further that colorectal cancer etiology differs by tumor MSI status.

What is microsatellite instability?

Microsatellites are repeated sequences of DNA.  The normal length of microsatellites in an individual’s cells are set at birth, although lengths vary from one person to another.  However, during the many divisions cells undergo in a person’s lifetime, mistakes can be made duplicating DNA which don’t get repaired, so microsatellites change in length in some tissues.  The presence of abnormally short or long microsatellites indicates that genes that should be repairing DNA are mutated and aren’t doing their job.

Mutations in DNA repair genes can lead to a particular form of colorectal cancer linked to microsatellite instability.  About 1 in 6 or 7 (15 percent) colorectal cancers are microsatellite instable. Some people are born with mutations in DNA repair genes, as in Lynch syndrome.  Other acquire mutations during their lives.

SOURCE: Campbell et al., Journal of the National Cancer Institute, Volume 102, Number 6, pages 391-400, March 17, 2010.

Briefly This Week

• Acupuncture didn’t reduce prolonged post-operative ileus.
• Large numbers of retrieved lymph nodes predict MSI in stage I and II colon cancer.
• Sarcomas found in families with Lynch syndrome.
• Changes in the weekly briefs.

Research News

Acupuncture doesn’t help prevent prolonged ileus after surgery

A collaboration between M.D. Anderson Cancer Center and Fudan University Cancer Hospital found that daily acupuncture did not reduce the number of patients who didn’t have bowel function return by the fourth day after their operation, a condition known as prolonged post-operative ileus (PPOI).  Bowel function included passing gas (flatus) or a bowel movement.

Patients were randomly assigned to a daily acupuncture treatment by an experienced physician acupuncturist.  By the fourth day, there was no difference in patients who had acupuncture and those who didn’t in either return of bowel function or quality of life.

In the January 7, 2010 issue of the World Journal of Gastroenterology, the study team concluded,

Acupuncture did not prevent prolonged post-operative ileus (PPOI) and was not useful for treating PPOI once it had developed in this population.

Return to top.

Many retrieved lymph nodes are strongly connected with microsatellite instability (MSI) in stage I and II colon cancer

Among 82 patients with stage I or II cancer, the average number of negative nodes found was 13.7 in a group of 71 with microsatellite stable tumors.  However, the average number for MSI tumors was 23.6.

The average number of examined nodes for all stage I and II cancers in the hospital where the study was completed was 15, and 9 of the 11 MSI tumors (82 percent) had more than this number of nodes retrieved and tested.

Writing in the January 2010 issue of Archives of Surgery, Clarisse Eveno, MD and her team said,

Good prognosis that is usually associated with tumors having a high number of uninvolved lymph nodes might reflect the high prevalence of MSI among these tumors. The number of examined lymph nodes as a quality criterion should be used with caution. For stage I or stage II colorectal carcinomas, restricting MSI phenotyping to tumors with more than the mean number of lymph nodes identifies almost all MSI tumors.

Return to top.

Sarcomas may be part of the Lynch syndrome

Although not usually considered part of the Lynch syndrome associated cancers, sarcomas have been identified in some families with the mutations.  Fourteen sarcomas were found among Lynch families with mutations in the MSH2, MSH6, and MLH1 genes in the Danish HNPCC registry. While most were in soft tissue, there were also uterine sarcomas.  The median age at diagnosis was 43, but cancers were diagnosed as young as 15 and as old as 74.

Reporting their analysis in Familial Cancer, September 2009, Mef Nilbert and the Danish team wrote,

This suggests that sarcomas may be part of the HNPCC tumor spectrum and that colorectal cancer should be considered in the family history of sarcoma patients.

Return to top.

What’s with the changes?

We’re experimenting with a little different format for the weekly briefs. You’ll find two different blogs — one with research news and another with upcoming events and consumer information. In this way, we hope to include a bit more detail while still keeping the individuals items brief and easy to read.

Your comments are welcome!

Return to top.

Watch for Kate Murphy on Twitter this weekend.  She’ll be sending Tweets from the ASCO GI Symposium in Orlando.  Follow C3 news and research updates @FIGHTCRC.   Here’s more information on joining Twitter to follow C3.

However, a new analysis of patients treated with FOLFOX (oxaliplatin, leucovorin, and 5-FU) found no differences between patients with deficient mismatch repair tumors and those with normal gene expression.

In this small study of 135 patients, the research team concluded that adding oxaliplatin to 5-FU and leucovorin may overcome resistance to chemotherapy in mismatch repair deficient and microsatellite instable (MSI) colon cancer.

South Korean researchers analyzed tumors from patients with stage II, III, or IV colorectal cancer, all of whom had surgery that completely removed all visible signs of cancer.  The 14 patients with stage IV cancer had liver metastases only, which were successfully removed.

They studied two genes related to deficient DNA repair — MLH1 and MSH2.  They also used genotyping to determine microsatellite instability (MSI), and stained tumors for the expression of p53 protein.

They found that:

• About 1 in 10  tumors were MSI-high (MSI-H); 9 out of 10 were MSI-low or microsatellite stable (MSI-L/S).
• Similarly, almost 10 percent showed deficient mismatch repair gene expression (MMR-D).
• There was almost complete agreement (94.7 percent) between the MSI-high and deficient mismatch repair specimens.

Comparing results to disease-free and overall survival, the study showed:

• There was no difference between patients with deficient and intact DNA mismatch repair for either disease-free or overall survival.
• p53 expression made no difference in either disease-free or overall survival.
• For patients with agreement between genotyping and immunohistochemical tests, microsatellite instability — either high or low/stable — also made no difference in disease-free or overall survival.

Seung Tae Kim and the team in Seoul concluded,

The MMR status or p53 positivity was not significantly associated with outcomes to FOLFOX as adjuvant chemotherapy in colon cancer patients with R0 resection. Adding oxaliplatin in adjuvant chemotherapy may overcome negative impact of 5-FU on colon cancers with MSI-H/MMR-D.

SOURCE: Kim et al., Cancer Chemotherapy and Pharmacology,Online First, December 24, 2009.

Patients whose cancers had high microsatellite instability (MSI) had significantly better outcomes at every stage, but mutations in the KRAS gene predicted poorer survival.

Scientists in New York cancer centers assessed MSI and KRAS genetic mutations in 532 primary colorectal cancers removed during surgery.  Twelve percent of cancer had high levels of microsatellite instability (MSI), while 36 percent had mutations in the KRAS gene.

MSI was more common in early stages with very little MSI in cancers diagnosed at stage IV where cancer had spread beyond the colon:

• stage I– 15 percent
• stage II — 21 percent
• stage III — 10 percent
• stage IV — 2 percent

KRAS mutations were more evenly distributed across stages:

• stage I– 36 percent
• stage II — 34 percent
• stage III — 35 percent
• stage IV — 40 percent

Patients with MSI were much less likely to die of cancer within five years of their diagnosis.  More than 9 out of ten (92 percent) were alive five years later compared to 6 of 10 (59 percent) of those who didn’t show MSI.

KRAS was the opposite story.  Mutations in KRAS led to poorer five year survival with 55 percent alive compared to 68 percent of patients with normal KRAS (wild-type).

The researchers noted that there was a group of stage I and II patients who had a particularly poor chance of living 5 years.  Those patients didn’t have MSI but did have KRAS mutations.

Garrett M. Nash from the Department of Surgery at Memorial Sloan Kettering and his colleagues concluded,

MSI and KRAS mutation provide fundamental genetic signatures influencing tumor behavior across patient subsets and stages of tumor development.

SOURCE: Nash et al., Annals of Surgical Oncology, published online October 8, 2009.

About 15 percent of colon cancers develop when damaged DNA is not repaired and mutated cells grow into malignant tumors.  So-called deficient mismatch repair (dMMR) tumors have features different from most colorectal cancer, including a better prognosis.  They also have a very poor response to 5-FU-based chemotherapy.

However, researchers studying tumor tissue from patients enrolled in a clinical trial comparing 5-FU and leucovorin alone to 5-FU, leucovorin, and irinotecan found that those with deficient mismatch repair tumors who received irinotecan had better disease-free survival and overall survival at five years than patients whose mismatch repair genes were working.  Those with dMMR on the 5-FU-only arm of the trial had no similar benefit.

Deficient mismatch repair can be identified by measuring  mutated microsatellites — short lengths of DNA that are abnormally shorter or longer because they were not caught and corrected during cell division.  This is known as microsatellite instability or MSI. dMMR can also be diagnosed by looking for missing mismatch repair gene expression in tumor tissue.

To find out if there was any benefit for irinotecan in deficient mismatch repair tumors, researchers tested tissue saved after surgery for both MSI and loss of MLH1 and MSH2 gene expression from patients enrolled in the CALGB-89803 clinical trial.

CALGB-89803 randomly assigned stage III colon cancer patients after surgery to chemotherapy to an IV shot of 5-FU plus intravenous leucovorin (FU/LV) or the same 5-FU and leucovorin treatment plus irinotecan (IFL).   Chemotherapy was given every week.

For all of the nearly 1,300 people enrolled in the trial, there was no significant difference in overall survival after 5 years (70 percent for IFL and 72 percent for FU/LV).  There was a significant increase in toxic side effects in the IFL arm and some unexpected deaths.

However, when tumors were analyzed by deficient mismatch repair status (dMMR), as evidenced by high microsatellite instability (MSI-H) compared to intact mismatch repair (iMMR) or low or stable microsatellite stability (MSI-L/S), adding irinotecan did make a difference in disease free survival and overall survival five years after surgery:

Five year disease free survival

• All patients:  61 percent
• FU/LV and MMR-D (MSI-high): 57 percent
• FU/LV and MMR-I (MSI low or stable): 51 percent   (no significant difference)
• IFL and MMR-I: 59 percent
• IFL and MMR-D: 76 percent  (significant difference in this group)

Overall survival at five years

• All patients:  67 percent
• FU/LV MMR-D:  67 percent
• FU/LV MMR-I:   72 percent
• IFL  MMR-I:   70 percent
• IFL MMR-D: 78 percent  (significant difference in this group of patients)

Some, but not all, dMMR tumors are caused by inherited genetic mutations in mismatch repair genes.  Lynch syndrome or hereditary non-polyposis colon cancer (HNPCC) leads to colorectal, uterine, and other cancers through such inherited mutations.

Since colon cancer caused by deficient mismatch repair does not benefit from 5-FU chemotherapy, it is common to test for MSI or missing MLH1 or MSH2 gene expression in tissue removed during surgery before prescribing chemotherapy.  Testing tissue also screens for Lynch syndrome.

Monica M. Bertagnolli, M.D. and her colleague in the Cancer and Leukemia Group B (CALGB) concluded,

Loss of tumor MMR function may predict improved outcome in patients treated with the IFL regimen as compared with those receiving FU/LV.

SOURCE:  Bertagnolli et al., Journal of Clinical Oncology, Volume 17, Number 11, April 10, 2009.

Last year at ASCO, Dr. Daniel Sargent presented new data that patients with stage II disease with microsatellite instability do not only not benefit from 5-FU, but they may be harmed, and it was recommended to test for MSI in all stage II colon cancer patients and in the presence of MSI-high not to give 5-FU. For stage III colon cancer that was not the case.

This year, the PETACC-3 clinical trial was analyzed for MSI and did not show the same the same findings. It seems that chemotherapy does not harm these patients, and they may benefit.

This has been an ongoing controversy over the last couple of years with some studies showing benefit and other not. Last year’s ASCO showed there even may be harm.  What MSI means is now again up in the air. We can certainly state that the presence of MSI is a GOOD prognostic marker, meaning that these patients have a lower risk of tumor recurrence. However, if chemotherapy is beneficial or not is still not clearly answered.

Another finding in this clinical trial showed that 18q deletions are not prognostic in stage II disease when MSI status is known. That is important because our clinical trial E-5202 in the US assumed that patients with an 18q deletion are at higher risk for tumor recurrence independent of MSI, which may alter the interpretation of the clinical trial.

All these data show that we are learning a tremendous amount about the molecular make up of tumors, but it also shows that it is not easy to develop clinically meaningful markers. However, there is no doubt that new markers will be identified and validated over the years to come and will make our personalized oncology care a reality.

Microsatellite instability has developed into the most important prognostic and predictive marker for patients with stage II colon cancer. Recent studies presented by Dr. Daniel Sargent at ASCO showed that patients with stage II colon cancer who have microsatellite instability did not benefit from 5-FU chemotherapy. Therefore these patients with no risk factors such as clinical obstruction, lymphovascular invasion, poorly differentiated adenocarcinoma, or insufficient lymph node collection should not receive 5-FU.

Since this data emerged we have evaluated every patients with stage II colon cancer for MSI to make a better treatment plan. MSI findings do not impact stage III disease (meaning positive lymph node involvement).

In the future it will  be more and more important to understand colon cancer better, including how it behaves despite looking the same under the microscope.  We don’t yet understand why MSI is associated with better outcomes, but colon cancers with MSI do not benefit from 5-FU chemotherapy.

There is, in fact a prospective clinical trial, ongoing called E-5202 which enrolls patients with stage II colon cancer who will be tested for MSI and 18q deletion. Patients with loss of 18q or microsatelite stable tumor will be randomly treated with FOLFOX with or without Avastin. Patients with tumors who did not lose 18q and have MSI will be only observed. This is an important trial to demonstrate if these clinical markers are helpful in treatment decisions.

Please make sure that you discuss these molecular markers when you have stage II colon cancer.