User login
Dear colleagues,
We are all often asked by friends, colleagues, and especially patients how to reduce the risk of getting colorectal cancer. We offer exercise, diet, and smoking cessation as some possible ways to mitigate risk. But what about that wonder drug – the ubiquitous aspirin? The American Gastroenterological Association’s recent clinical practice update suggests that aspirin may be protective in some patients younger than 70 years depending on their cardiovascular and gastrointestinal bleeding risks. If so, should we gastroenterologists be the ones to recommend or even prescribe aspirin? Or are the data just not there yet? We invite two colorectal cancer experts, Dr. Sonia Kupfer and Dr. Jennifer Weiss, to share their perspectives in light of these new recommendations. I invite you to a great debate and look forward to hearing your own thoughts online and by email at [email protected].
Gyanprakash A. Ketwaroo, MD, MSc, is assistant professor of medicine at Baylor College of Medicine, Houston. He is an associate editor for GI & Hepatology News.
Not our lane
By Jennifer Weiss, MD, MS
In 2021, the AGA published a clinical practice update on chemoprevention for colorectal neoplasia that advises clinicians to use low-dose aspirin to reduce colorectal cancer (CRC) incidence and mortality in average-risk individuals who are (1) younger than 70 years with a life expectancy of at least 10 years, (2) have at least a 10% 10-year cardiovascular disease (CVD) risk, and (3) are not at high risk for gastrointestinal bleeding.1 As gastroenterologists, we may see average-risk patients only at the time of their screening or surveillance colonoscopies, and I wonder if we should be taking the lead in prescribing/recommending aspirin for CRC chemoprevention in these patients. To answer this question, I will review three main concerns: (1) issues with the overall strength of the evidence on the effectiveness of aspirin to reduce CRC incidence and mortality, (2) determining an individual’s long-term CVD risk and life expectancy may be outside of a gastroenterologist’s purview, and (3) the potential for serious gastrointestinal bleeding is dynamic and requires continual review.
Studies examining the effects of aspirin on CRC incidence and mortality have limitations and mixed results. Many of the randomized controlled trials have primarily been secondary analyses of studies with primary CVD endpoints. When examined individually some studies show no significant reduction in CRC risk such as the Women’s Health Study (at 10 years of follow-up), the Swedish Aspirin Low-Dose Trial, and the UK-TIA Aspirin Trial, while some meta-analyses have shown a decrease in CRC incidence and mortality.2 One reason for this discrepancy may be varying lengths of follow-up across studies. In addition, we do not yet know the optimal aspirin dose or duration of therapy. The protective effect of aspirin on CRC incidence and mortality in average-risk individuals is mostly seen after 10-20 years of follow-up. This is relevant to the first part of the AGA clinical practice update recommendation that refers to individuals with a life expectancy of at least 10 years. The second part of the recommendation includes individuals with a 10-year CVD risk of at least 10%. As gastroenterologists, we may see these patients only two to three times over a 10-20 year period and only for their screening/surveillance colonoscopy. I would argue that we are not in the best position to address changes in life expectancy and 10-year CVD risk status over time and determine if they should start or continue taking aspirin for CRC chemoprevention.
The United States Preventive Services Task Force is also reexamining their previous recommendations for aspirin for primary prevention of cardiovascular disease. The 2016 guidelines recommended initiation of low-dose aspirin for primary prevention of CVD and CRC in adults aged 50-59 years who have a 10% or greater 10-year CVD risk and at least a 10-year life expectancy (Grade B). The current draft recommendations state that aspirin use for the primary prevention of CVD events in adults aged 40-59 years who have a 10% or greater 10-year CVD risk has a small net benefit (Grade C) and that initiating aspirin for the primary prevention of CVD events in adults aged 60 years and older has no net benefit (Grade D). They also state that, based on longer-term follow-up data from the Women’s Health Study and newer trials, the evidence is inadequate that low-dose aspirin use reduces CRC incidence and mortality.3 Because of these moving targets, we may also find ourselves walking back the AGA clinical practice update recommendations in the future.
One main concern for long-term aspirin use is the potential for gastrointestinal bleeding. Participants in more than one of the CVD prevention trials had a significant increase in gastrointestinal bleeding.1,2 While gastrointestinal bleeding falls within our wheelhouse, we are not always privy to a patient’s risk factors for bleeding. For example, patients may receive multiple courses of steroids for arthritis or chronic pulmonary disorders and not take concomitant acid suppression. These risks are dynamic and require continual reassessment as individuals age, new diagnoses are made, and new medications are started or stopped by providers other than their gastroenterologist. If a patient is taking aspirin, regardless of the reason, we need to make sure it is correctly recorded in their medication list, especially if they are obtaining it over the counter. This is one area where we should definitely play a role.
There is a population in which I do recommend aspirin for reduction of CRC chemoprevention – individuals with Lynch syndrome. I believe the data for the protective effects of aspirin on CRC incidence are much stronger for individuals with Lynch syndrome than the average-risk population. The CAPP2 trial was a randomized trial with a two-by-two factorial design where individuals with Lynch syndrome were randomly assigned to aspirin 600 mg/day or aspirin placebo or resistant starch or starch placebo for up to 4 years. The primary endpoint of this trial was development of CRC (unlike the CVD trials referred to earlier in this article). Long-term follow-up of the CAPP2 trial participants found a significantly decreased risk of CRC after 2 years of aspirin use (hazard ratio, 0.56, 95% confidence interval, 0.34-0.91).4 The current CAPP3 trial will answer questions about the effectiveness of lower doses of aspirin (100 mg and 300 mg).
The recommendation for aspirin use for CRC chemoprevention in average-risk individuals depends on multiple factors (life expectancy, determination of CVD risk, and dynamic assessment of gastrointestinal bleeding risk) that are outside the purview of a gastroenterologist who sees the patient only at a screening or surveillance colonoscopy. This is not in our lane. What is in our lane, however, is the recommendation for aspirin use for CRC chemoprevention in select high-risk populations such as individuals with Lynch syndrome.
Dr. Weiss is associate professor in the division of gastroenterology and hepatology and director of the University of Wisconsin Gastroenterology Genetics Clinic at University of Wisconsin School of Medicine and Public Health. She reports receiving research support from Exact Sciences as a site-PI of a multisite trial.
References
1. Liang PS et al. Clin Gastroenterol Hepatol. 2021 Jul;19(7):1327-36. doi: 10.1016/j.cgh.2021.02.014
2. Katona BW and Weiss JM. Gastroenterology. 2020 Jan;158(2):368-88. doi: 10.1053/j.gastro.2019.06.047
3. United States Preventive Services Task Force. “Aspirin Use to Prevent Cardiovascular Disease: Preventive Medication.” Accessed April 5, 2022.
4. Burn J et al. Lancet. 2020 Jun13;395(10240):1855-63. doi: 10.1016/s0140-6736(20)30366-4
Yes, but individualize it
By Sonia S. Kupfer, MD
Colorectal cancer (CRC) is one of the top three causes of cancer and cancer death worldwide with an alarming rise in younger adults. Preventive strategies including screening, chemoprevention, and risk factor modification are important to reduce overall CRC burden. Aspirin, which is cheap and readily available, is supported for CRC chemoprevention by multiple lines of strong evidence. Recent AGA practice guidelines recommend low-dose aspirin chemoprevention in individuals at average CRC risk who are younger than 70 years with a life expectancy of at least 10 years, have a 10-year cardiovascular disease risk of at least 10% and are not at high risk for bleeding.1 This advice diverges from the most recent U.S. Preventive Services Task Force–proposed guidelines2 that reverse the 2016 USPSTF recommendation for aspirin CRC chemoprevention (and primary prevention of cardiovascular disease) based on uncertainty of net benefit over harms, especially in older individuals. In light of conflicting advice, how should we counsel our patients about aspirin use for CRC chemoprevention? In my opinion, we shouldn’t “throw the baby out with the bathwater” and should follow the AGA practice guideline to individualize aspirin chemoprevention based on balancing known benefits and risks.
As reviewed in the AGA practice guidelines1, many, but not all, randomized controlled and observational trials have shown efficacy of aspirin for reduction of CRC mortality, incidence, and adenoma recurrence. Analysis of cardiovascular prevention trials including over 14,000 mostly middle-aged people showed 33% reduction in 20-year cumulative CRC mortality. While a pooled estimate of four trials did not show reduced incidence 0-12 years after aspirin initiation, as noted in the practice guideline, three of these trials did show a 40% reduction between 10 and 19 years, a finding that is in line with results from a 20-year pooled analysis showing 24% reduction in CRC incidence by aspirin. Among Lynch syndrome patients, exposure to high-dose aspirin also significantly reduced CRC incidence in a randomized controlled trial with up to 20 years of follow-up3 highlighting that chemoprotective effects take years to manifest, and long-term follow-up in cancer chemoprevention trials is needed. Studies also have shown reduced adenoma incidence or recurrence by aspirin ranging from 17% to 51% depending on the study population, dose, and adherence. In addition to clinical trials, experimental data have demonstrated protective cellular effects of aspirin on colonic carcinogenesis, though exact mechanisms of this protective effect remain incompletely understood and are active areas of research, including in my lab. Taken together, there is a large body of evidence supporting a protective effect of aspirin on CRC mortality and colorectal neoplasia incidence most evident after 1-2 decades of follow-up.
Not all trials have shown that aspirin is chemoprotective, and, in fact, the ASPREE trial,4 that randomized over 19,000 healthy adults over the age of 70 to 100 mg aspirin or placebo, showed increased cancer mortality when the trial was stopped prematurely after 5 years. Individuals who started aspirin under age 70 appear to have continued chemoprotection as they age5 suggesting that aspirin, if it is tolerated, might not necessarily need to be stopped at a certain age. Notably, the ASPREE trial did not show increased CRC incidence, which begs the question of the biological mechanism underlying increased cancer mortality in trial participants. Beyond the findings of ASPREE, aspirin use is associated with risks of intracranial and gastrointestinal bleeding with estimated odds ratios of 1.29 and 1.59, respectively. The AGA practice guideline acknowledges these risks especially in older adults and recommends initiation of aspirin in individuals under the age of 70 who are expected to live another 10 years without bleeding risks in order to reap the benefits and minimize the risks.
Risk stratification hinges on acceptance and feasibility. Three-quarters of providers, when surveyed, reported aspirin to be a suitable preventive treatment with more favorable views expressed by gastroenterologists and genetics providers, compared with colorectal surgeons.6 In Lynch syndrome, rates of aspirin chemoprevention recommendation by providers in real-world practices ranged from 35% to 67%; my own practice strives to discuss aspirin use with every Lynch patient at every clinic and endoscopy visit. Real-world data for uptake and adherence of aspirin CRC chemoprevention are sparse. Uptake and adherence of aspirin for cancer chemoprevention in clinical trials ranged from 41% to 80% with good adherence, although these findings likely are not generalizable to routine practice. Current blood pressure and cholesterol guidelines for primary prevention include calculation of 10-year cardiovascular risk using automatic calculators in the electronic health record; thus, it should be relatively straightforward to apply this approach for aspirin CRC chemoprevention as well. While calculation of bleeding risk is less well established, there are publicly available calculators that combine cardiovascular and bleeding risk for primary aspirin prevention and such decision aids should be explored for aspirin CRC chemoprevention. However, given the recent recommendation reversal by the USPSTF, I am concerned that recommendation and uptake of aspirin CRC chemoprevention will decline substantially.
In order to reduce CRC burden, we should employ everything in our armamentarium including aspirin chemoprevention. Individualized risk assessment for aspirin chemoprevention, as advised by the AGA practice guideline, will enable the right people to benefit while minimizing risks. Future studies should strengthen the evidence base for aspirin CRC chemoprevention and refine risk stratification, including for younger individuals given the rise in early-onset CRC. The optimal approach to aspirin chemoprevention was best summed up by the foremost expert in the field, Dr. Andy Chan, to the New York Times:7 “we need to think about personalizing who we give aspirin to, and move away from a one-size-fits-all solution”.
Dr. Kupfer is associate professor of medicine, director of the Gastrointestinal Cancer Risk and Prevention Clinic, and codirector of the Comprehensive Cancer Risk and Prevention Clinic at the University of Chicago. She reports no relevant conflicts of interest.
References
1. Liang PS et al. Clin Gastroenterol Hepatol. 2021 Jul;19(7):1327-36.
2. United States Preventive Services Task Force. “Aspirin Use to Prevent Cardiovascular Disease: Preventive Medication.” Accessed April 10, 2022.
3. Burn J et al. Lancet. 2020 Jun 13;395(10240):1855-63.
4. McNeil JJ et al. N Engl J Med. 2018 Oct 18;379(16):1519-28.
5. Guo CG et al. JAMA Oncol. 2021 Mar 1;7(3):428-35.
6. Lloyd KE et al. Prev Med. 2022 Jan;154:106872.
7. Rabin RC. “Aspirin Use to Prevent 1st Heart Attack or Stroke Should Be Curtailed, U.S. Panel Says.” New York Times. Oct. 13, 2021. Accessed April 10, 2022.
Dear colleagues,
We are all often asked by friends, colleagues, and especially patients how to reduce the risk of getting colorectal cancer. We offer exercise, diet, and smoking cessation as some possible ways to mitigate risk. But what about that wonder drug – the ubiquitous aspirin? The American Gastroenterological Association’s recent clinical practice update suggests that aspirin may be protective in some patients younger than 70 years depending on their cardiovascular and gastrointestinal bleeding risks. If so, should we gastroenterologists be the ones to recommend or even prescribe aspirin? Or are the data just not there yet? We invite two colorectal cancer experts, Dr. Sonia Kupfer and Dr. Jennifer Weiss, to share their perspectives in light of these new recommendations. I invite you to a great debate and look forward to hearing your own thoughts online and by email at [email protected].
Gyanprakash A. Ketwaroo, MD, MSc, is assistant professor of medicine at Baylor College of Medicine, Houston. He is an associate editor for GI & Hepatology News.
Not our lane
By Jennifer Weiss, MD, MS
In 2021, the AGA published a clinical practice update on chemoprevention for colorectal neoplasia that advises clinicians to use low-dose aspirin to reduce colorectal cancer (CRC) incidence and mortality in average-risk individuals who are (1) younger than 70 years with a life expectancy of at least 10 years, (2) have at least a 10% 10-year cardiovascular disease (CVD) risk, and (3) are not at high risk for gastrointestinal bleeding.1 As gastroenterologists, we may see average-risk patients only at the time of their screening or surveillance colonoscopies, and I wonder if we should be taking the lead in prescribing/recommending aspirin for CRC chemoprevention in these patients. To answer this question, I will review three main concerns: (1) issues with the overall strength of the evidence on the effectiveness of aspirin to reduce CRC incidence and mortality, (2) determining an individual’s long-term CVD risk and life expectancy may be outside of a gastroenterologist’s purview, and (3) the potential for serious gastrointestinal bleeding is dynamic and requires continual review.
Studies examining the effects of aspirin on CRC incidence and mortality have limitations and mixed results. Many of the randomized controlled trials have primarily been secondary analyses of studies with primary CVD endpoints. When examined individually some studies show no significant reduction in CRC risk such as the Women’s Health Study (at 10 years of follow-up), the Swedish Aspirin Low-Dose Trial, and the UK-TIA Aspirin Trial, while some meta-analyses have shown a decrease in CRC incidence and mortality.2 One reason for this discrepancy may be varying lengths of follow-up across studies. In addition, we do not yet know the optimal aspirin dose or duration of therapy. The protective effect of aspirin on CRC incidence and mortality in average-risk individuals is mostly seen after 10-20 years of follow-up. This is relevant to the first part of the AGA clinical practice update recommendation that refers to individuals with a life expectancy of at least 10 years. The second part of the recommendation includes individuals with a 10-year CVD risk of at least 10%. As gastroenterologists, we may see these patients only two to three times over a 10-20 year period and only for their screening/surveillance colonoscopy. I would argue that we are not in the best position to address changes in life expectancy and 10-year CVD risk status over time and determine if they should start or continue taking aspirin for CRC chemoprevention.
The United States Preventive Services Task Force is also reexamining their previous recommendations for aspirin for primary prevention of cardiovascular disease. The 2016 guidelines recommended initiation of low-dose aspirin for primary prevention of CVD and CRC in adults aged 50-59 years who have a 10% or greater 10-year CVD risk and at least a 10-year life expectancy (Grade B). The current draft recommendations state that aspirin use for the primary prevention of CVD events in adults aged 40-59 years who have a 10% or greater 10-year CVD risk has a small net benefit (Grade C) and that initiating aspirin for the primary prevention of CVD events in adults aged 60 years and older has no net benefit (Grade D). They also state that, based on longer-term follow-up data from the Women’s Health Study and newer trials, the evidence is inadequate that low-dose aspirin use reduces CRC incidence and mortality.3 Because of these moving targets, we may also find ourselves walking back the AGA clinical practice update recommendations in the future.
One main concern for long-term aspirin use is the potential for gastrointestinal bleeding. Participants in more than one of the CVD prevention trials had a significant increase in gastrointestinal bleeding.1,2 While gastrointestinal bleeding falls within our wheelhouse, we are not always privy to a patient’s risk factors for bleeding. For example, patients may receive multiple courses of steroids for arthritis or chronic pulmonary disorders and not take concomitant acid suppression. These risks are dynamic and require continual reassessment as individuals age, new diagnoses are made, and new medications are started or stopped by providers other than their gastroenterologist. If a patient is taking aspirin, regardless of the reason, we need to make sure it is correctly recorded in their medication list, especially if they are obtaining it over the counter. This is one area where we should definitely play a role.
There is a population in which I do recommend aspirin for reduction of CRC chemoprevention – individuals with Lynch syndrome. I believe the data for the protective effects of aspirin on CRC incidence are much stronger for individuals with Lynch syndrome than the average-risk population. The CAPP2 trial was a randomized trial with a two-by-two factorial design where individuals with Lynch syndrome were randomly assigned to aspirin 600 mg/day or aspirin placebo or resistant starch or starch placebo for up to 4 years. The primary endpoint of this trial was development of CRC (unlike the CVD trials referred to earlier in this article). Long-term follow-up of the CAPP2 trial participants found a significantly decreased risk of CRC after 2 years of aspirin use (hazard ratio, 0.56, 95% confidence interval, 0.34-0.91).4 The current CAPP3 trial will answer questions about the effectiveness of lower doses of aspirin (100 mg and 300 mg).
The recommendation for aspirin use for CRC chemoprevention in average-risk individuals depends on multiple factors (life expectancy, determination of CVD risk, and dynamic assessment of gastrointestinal bleeding risk) that are outside the purview of a gastroenterologist who sees the patient only at a screening or surveillance colonoscopy. This is not in our lane. What is in our lane, however, is the recommendation for aspirin use for CRC chemoprevention in select high-risk populations such as individuals with Lynch syndrome.
Dr. Weiss is associate professor in the division of gastroenterology and hepatology and director of the University of Wisconsin Gastroenterology Genetics Clinic at University of Wisconsin School of Medicine and Public Health. She reports receiving research support from Exact Sciences as a site-PI of a multisite trial.
References
1. Liang PS et al. Clin Gastroenterol Hepatol. 2021 Jul;19(7):1327-36. doi: 10.1016/j.cgh.2021.02.014
2. Katona BW and Weiss JM. Gastroenterology. 2020 Jan;158(2):368-88. doi: 10.1053/j.gastro.2019.06.047
3. United States Preventive Services Task Force. “Aspirin Use to Prevent Cardiovascular Disease: Preventive Medication.” Accessed April 5, 2022.
4. Burn J et al. Lancet. 2020 Jun13;395(10240):1855-63. doi: 10.1016/s0140-6736(20)30366-4
Yes, but individualize it
By Sonia S. Kupfer, MD
Colorectal cancer (CRC) is one of the top three causes of cancer and cancer death worldwide with an alarming rise in younger adults. Preventive strategies including screening, chemoprevention, and risk factor modification are important to reduce overall CRC burden. Aspirin, which is cheap and readily available, is supported for CRC chemoprevention by multiple lines of strong evidence. Recent AGA practice guidelines recommend low-dose aspirin chemoprevention in individuals at average CRC risk who are younger than 70 years with a life expectancy of at least 10 years, have a 10-year cardiovascular disease risk of at least 10% and are not at high risk for bleeding.1 This advice diverges from the most recent U.S. Preventive Services Task Force–proposed guidelines2 that reverse the 2016 USPSTF recommendation for aspirin CRC chemoprevention (and primary prevention of cardiovascular disease) based on uncertainty of net benefit over harms, especially in older individuals. In light of conflicting advice, how should we counsel our patients about aspirin use for CRC chemoprevention? In my opinion, we shouldn’t “throw the baby out with the bathwater” and should follow the AGA practice guideline to individualize aspirin chemoprevention based on balancing known benefits and risks.
As reviewed in the AGA practice guidelines1, many, but not all, randomized controlled and observational trials have shown efficacy of aspirin for reduction of CRC mortality, incidence, and adenoma recurrence. Analysis of cardiovascular prevention trials including over 14,000 mostly middle-aged people showed 33% reduction in 20-year cumulative CRC mortality. While a pooled estimate of four trials did not show reduced incidence 0-12 years after aspirin initiation, as noted in the practice guideline, three of these trials did show a 40% reduction between 10 and 19 years, a finding that is in line with results from a 20-year pooled analysis showing 24% reduction in CRC incidence by aspirin. Among Lynch syndrome patients, exposure to high-dose aspirin also significantly reduced CRC incidence in a randomized controlled trial with up to 20 years of follow-up3 highlighting that chemoprotective effects take years to manifest, and long-term follow-up in cancer chemoprevention trials is needed. Studies also have shown reduced adenoma incidence or recurrence by aspirin ranging from 17% to 51% depending on the study population, dose, and adherence. In addition to clinical trials, experimental data have demonstrated protective cellular effects of aspirin on colonic carcinogenesis, though exact mechanisms of this protective effect remain incompletely understood and are active areas of research, including in my lab. Taken together, there is a large body of evidence supporting a protective effect of aspirin on CRC mortality and colorectal neoplasia incidence most evident after 1-2 decades of follow-up.
Not all trials have shown that aspirin is chemoprotective, and, in fact, the ASPREE trial,4 that randomized over 19,000 healthy adults over the age of 70 to 100 mg aspirin or placebo, showed increased cancer mortality when the trial was stopped prematurely after 5 years. Individuals who started aspirin under age 70 appear to have continued chemoprotection as they age5 suggesting that aspirin, if it is tolerated, might not necessarily need to be stopped at a certain age. Notably, the ASPREE trial did not show increased CRC incidence, which begs the question of the biological mechanism underlying increased cancer mortality in trial participants. Beyond the findings of ASPREE, aspirin use is associated with risks of intracranial and gastrointestinal bleeding with estimated odds ratios of 1.29 and 1.59, respectively. The AGA practice guideline acknowledges these risks especially in older adults and recommends initiation of aspirin in individuals under the age of 70 who are expected to live another 10 years without bleeding risks in order to reap the benefits and minimize the risks.
Risk stratification hinges on acceptance and feasibility. Three-quarters of providers, when surveyed, reported aspirin to be a suitable preventive treatment with more favorable views expressed by gastroenterologists and genetics providers, compared with colorectal surgeons.6 In Lynch syndrome, rates of aspirin chemoprevention recommendation by providers in real-world practices ranged from 35% to 67%; my own practice strives to discuss aspirin use with every Lynch patient at every clinic and endoscopy visit. Real-world data for uptake and adherence of aspirin CRC chemoprevention are sparse. Uptake and adherence of aspirin for cancer chemoprevention in clinical trials ranged from 41% to 80% with good adherence, although these findings likely are not generalizable to routine practice. Current blood pressure and cholesterol guidelines for primary prevention include calculation of 10-year cardiovascular risk using automatic calculators in the electronic health record; thus, it should be relatively straightforward to apply this approach for aspirin CRC chemoprevention as well. While calculation of bleeding risk is less well established, there are publicly available calculators that combine cardiovascular and bleeding risk for primary aspirin prevention and such decision aids should be explored for aspirin CRC chemoprevention. However, given the recent recommendation reversal by the USPSTF, I am concerned that recommendation and uptake of aspirin CRC chemoprevention will decline substantially.
In order to reduce CRC burden, we should employ everything in our armamentarium including aspirin chemoprevention. Individualized risk assessment for aspirin chemoprevention, as advised by the AGA practice guideline, will enable the right people to benefit while minimizing risks. Future studies should strengthen the evidence base for aspirin CRC chemoprevention and refine risk stratification, including for younger individuals given the rise in early-onset CRC. The optimal approach to aspirin chemoprevention was best summed up by the foremost expert in the field, Dr. Andy Chan, to the New York Times:7 “we need to think about personalizing who we give aspirin to, and move away from a one-size-fits-all solution”.
Dr. Kupfer is associate professor of medicine, director of the Gastrointestinal Cancer Risk and Prevention Clinic, and codirector of the Comprehensive Cancer Risk and Prevention Clinic at the University of Chicago. She reports no relevant conflicts of interest.
References
1. Liang PS et al. Clin Gastroenterol Hepatol. 2021 Jul;19(7):1327-36.
2. United States Preventive Services Task Force. “Aspirin Use to Prevent Cardiovascular Disease: Preventive Medication.” Accessed April 10, 2022.
3. Burn J et al. Lancet. 2020 Jun 13;395(10240):1855-63.
4. McNeil JJ et al. N Engl J Med. 2018 Oct 18;379(16):1519-28.
5. Guo CG et al. JAMA Oncol. 2021 Mar 1;7(3):428-35.
6. Lloyd KE et al. Prev Med. 2022 Jan;154:106872.
7. Rabin RC. “Aspirin Use to Prevent 1st Heart Attack or Stroke Should Be Curtailed, U.S. Panel Says.” New York Times. Oct. 13, 2021. Accessed April 10, 2022.
Dear colleagues,
We are all often asked by friends, colleagues, and especially patients how to reduce the risk of getting colorectal cancer. We offer exercise, diet, and smoking cessation as some possible ways to mitigate risk. But what about that wonder drug – the ubiquitous aspirin? The American Gastroenterological Association’s recent clinical practice update suggests that aspirin may be protective in some patients younger than 70 years depending on their cardiovascular and gastrointestinal bleeding risks. If so, should we gastroenterologists be the ones to recommend or even prescribe aspirin? Or are the data just not there yet? We invite two colorectal cancer experts, Dr. Sonia Kupfer and Dr. Jennifer Weiss, to share their perspectives in light of these new recommendations. I invite you to a great debate and look forward to hearing your own thoughts online and by email at [email protected].
Gyanprakash A. Ketwaroo, MD, MSc, is assistant professor of medicine at Baylor College of Medicine, Houston. He is an associate editor for GI & Hepatology News.
Not our lane
By Jennifer Weiss, MD, MS
In 2021, the AGA published a clinical practice update on chemoprevention for colorectal neoplasia that advises clinicians to use low-dose aspirin to reduce colorectal cancer (CRC) incidence and mortality in average-risk individuals who are (1) younger than 70 years with a life expectancy of at least 10 years, (2) have at least a 10% 10-year cardiovascular disease (CVD) risk, and (3) are not at high risk for gastrointestinal bleeding.1 As gastroenterologists, we may see average-risk patients only at the time of their screening or surveillance colonoscopies, and I wonder if we should be taking the lead in prescribing/recommending aspirin for CRC chemoprevention in these patients. To answer this question, I will review three main concerns: (1) issues with the overall strength of the evidence on the effectiveness of aspirin to reduce CRC incidence and mortality, (2) determining an individual’s long-term CVD risk and life expectancy may be outside of a gastroenterologist’s purview, and (3) the potential for serious gastrointestinal bleeding is dynamic and requires continual review.
Studies examining the effects of aspirin on CRC incidence and mortality have limitations and mixed results. Many of the randomized controlled trials have primarily been secondary analyses of studies with primary CVD endpoints. When examined individually some studies show no significant reduction in CRC risk such as the Women’s Health Study (at 10 years of follow-up), the Swedish Aspirin Low-Dose Trial, and the UK-TIA Aspirin Trial, while some meta-analyses have shown a decrease in CRC incidence and mortality.2 One reason for this discrepancy may be varying lengths of follow-up across studies. In addition, we do not yet know the optimal aspirin dose or duration of therapy. The protective effect of aspirin on CRC incidence and mortality in average-risk individuals is mostly seen after 10-20 years of follow-up. This is relevant to the first part of the AGA clinical practice update recommendation that refers to individuals with a life expectancy of at least 10 years. The second part of the recommendation includes individuals with a 10-year CVD risk of at least 10%. As gastroenterologists, we may see these patients only two to three times over a 10-20 year period and only for their screening/surveillance colonoscopy. I would argue that we are not in the best position to address changes in life expectancy and 10-year CVD risk status over time and determine if they should start or continue taking aspirin for CRC chemoprevention.
The United States Preventive Services Task Force is also reexamining their previous recommendations for aspirin for primary prevention of cardiovascular disease. The 2016 guidelines recommended initiation of low-dose aspirin for primary prevention of CVD and CRC in adults aged 50-59 years who have a 10% or greater 10-year CVD risk and at least a 10-year life expectancy (Grade B). The current draft recommendations state that aspirin use for the primary prevention of CVD events in adults aged 40-59 years who have a 10% or greater 10-year CVD risk has a small net benefit (Grade C) and that initiating aspirin for the primary prevention of CVD events in adults aged 60 years and older has no net benefit (Grade D). They also state that, based on longer-term follow-up data from the Women’s Health Study and newer trials, the evidence is inadequate that low-dose aspirin use reduces CRC incidence and mortality.3 Because of these moving targets, we may also find ourselves walking back the AGA clinical practice update recommendations in the future.
One main concern for long-term aspirin use is the potential for gastrointestinal bleeding. Participants in more than one of the CVD prevention trials had a significant increase in gastrointestinal bleeding.1,2 While gastrointestinal bleeding falls within our wheelhouse, we are not always privy to a patient’s risk factors for bleeding. For example, patients may receive multiple courses of steroids for arthritis or chronic pulmonary disorders and not take concomitant acid suppression. These risks are dynamic and require continual reassessment as individuals age, new diagnoses are made, and new medications are started or stopped by providers other than their gastroenterologist. If a patient is taking aspirin, regardless of the reason, we need to make sure it is correctly recorded in their medication list, especially if they are obtaining it over the counter. This is one area where we should definitely play a role.
There is a population in which I do recommend aspirin for reduction of CRC chemoprevention – individuals with Lynch syndrome. I believe the data for the protective effects of aspirin on CRC incidence are much stronger for individuals with Lynch syndrome than the average-risk population. The CAPP2 trial was a randomized trial with a two-by-two factorial design where individuals with Lynch syndrome were randomly assigned to aspirin 600 mg/day or aspirin placebo or resistant starch or starch placebo for up to 4 years. The primary endpoint of this trial was development of CRC (unlike the CVD trials referred to earlier in this article). Long-term follow-up of the CAPP2 trial participants found a significantly decreased risk of CRC after 2 years of aspirin use (hazard ratio, 0.56, 95% confidence interval, 0.34-0.91).4 The current CAPP3 trial will answer questions about the effectiveness of lower doses of aspirin (100 mg and 300 mg).
The recommendation for aspirin use for CRC chemoprevention in average-risk individuals depends on multiple factors (life expectancy, determination of CVD risk, and dynamic assessment of gastrointestinal bleeding risk) that are outside the purview of a gastroenterologist who sees the patient only at a screening or surveillance colonoscopy. This is not in our lane. What is in our lane, however, is the recommendation for aspirin use for CRC chemoprevention in select high-risk populations such as individuals with Lynch syndrome.
Dr. Weiss is associate professor in the division of gastroenterology and hepatology and director of the University of Wisconsin Gastroenterology Genetics Clinic at University of Wisconsin School of Medicine and Public Health. She reports receiving research support from Exact Sciences as a site-PI of a multisite trial.
References
1. Liang PS et al. Clin Gastroenterol Hepatol. 2021 Jul;19(7):1327-36. doi: 10.1016/j.cgh.2021.02.014
2. Katona BW and Weiss JM. Gastroenterology. 2020 Jan;158(2):368-88. doi: 10.1053/j.gastro.2019.06.047
3. United States Preventive Services Task Force. “Aspirin Use to Prevent Cardiovascular Disease: Preventive Medication.” Accessed April 5, 2022.
4. Burn J et al. Lancet. 2020 Jun13;395(10240):1855-63. doi: 10.1016/s0140-6736(20)30366-4
Yes, but individualize it
By Sonia S. Kupfer, MD
Colorectal cancer (CRC) is one of the top three causes of cancer and cancer death worldwide with an alarming rise in younger adults. Preventive strategies including screening, chemoprevention, and risk factor modification are important to reduce overall CRC burden. Aspirin, which is cheap and readily available, is supported for CRC chemoprevention by multiple lines of strong evidence. Recent AGA practice guidelines recommend low-dose aspirin chemoprevention in individuals at average CRC risk who are younger than 70 years with a life expectancy of at least 10 years, have a 10-year cardiovascular disease risk of at least 10% and are not at high risk for bleeding.1 This advice diverges from the most recent U.S. Preventive Services Task Force–proposed guidelines2 that reverse the 2016 USPSTF recommendation for aspirin CRC chemoprevention (and primary prevention of cardiovascular disease) based on uncertainty of net benefit over harms, especially in older individuals. In light of conflicting advice, how should we counsel our patients about aspirin use for CRC chemoprevention? In my opinion, we shouldn’t “throw the baby out with the bathwater” and should follow the AGA practice guideline to individualize aspirin chemoprevention based on balancing known benefits and risks.
As reviewed in the AGA practice guidelines1, many, but not all, randomized controlled and observational trials have shown efficacy of aspirin for reduction of CRC mortality, incidence, and adenoma recurrence. Analysis of cardiovascular prevention trials including over 14,000 mostly middle-aged people showed 33% reduction in 20-year cumulative CRC mortality. While a pooled estimate of four trials did not show reduced incidence 0-12 years after aspirin initiation, as noted in the practice guideline, three of these trials did show a 40% reduction between 10 and 19 years, a finding that is in line with results from a 20-year pooled analysis showing 24% reduction in CRC incidence by aspirin. Among Lynch syndrome patients, exposure to high-dose aspirin also significantly reduced CRC incidence in a randomized controlled trial with up to 20 years of follow-up3 highlighting that chemoprotective effects take years to manifest, and long-term follow-up in cancer chemoprevention trials is needed. Studies also have shown reduced adenoma incidence or recurrence by aspirin ranging from 17% to 51% depending on the study population, dose, and adherence. In addition to clinical trials, experimental data have demonstrated protective cellular effects of aspirin on colonic carcinogenesis, though exact mechanisms of this protective effect remain incompletely understood and are active areas of research, including in my lab. Taken together, there is a large body of evidence supporting a protective effect of aspirin on CRC mortality and colorectal neoplasia incidence most evident after 1-2 decades of follow-up.
Not all trials have shown that aspirin is chemoprotective, and, in fact, the ASPREE trial,4 that randomized over 19,000 healthy adults over the age of 70 to 100 mg aspirin or placebo, showed increased cancer mortality when the trial was stopped prematurely after 5 years. Individuals who started aspirin under age 70 appear to have continued chemoprotection as they age5 suggesting that aspirin, if it is tolerated, might not necessarily need to be stopped at a certain age. Notably, the ASPREE trial did not show increased CRC incidence, which begs the question of the biological mechanism underlying increased cancer mortality in trial participants. Beyond the findings of ASPREE, aspirin use is associated with risks of intracranial and gastrointestinal bleeding with estimated odds ratios of 1.29 and 1.59, respectively. The AGA practice guideline acknowledges these risks especially in older adults and recommends initiation of aspirin in individuals under the age of 70 who are expected to live another 10 years without bleeding risks in order to reap the benefits and minimize the risks.
Risk stratification hinges on acceptance and feasibility. Three-quarters of providers, when surveyed, reported aspirin to be a suitable preventive treatment with more favorable views expressed by gastroenterologists and genetics providers, compared with colorectal surgeons.6 In Lynch syndrome, rates of aspirin chemoprevention recommendation by providers in real-world practices ranged from 35% to 67%; my own practice strives to discuss aspirin use with every Lynch patient at every clinic and endoscopy visit. Real-world data for uptake and adherence of aspirin CRC chemoprevention are sparse. Uptake and adherence of aspirin for cancer chemoprevention in clinical trials ranged from 41% to 80% with good adherence, although these findings likely are not generalizable to routine practice. Current blood pressure and cholesterol guidelines for primary prevention include calculation of 10-year cardiovascular risk using automatic calculators in the electronic health record; thus, it should be relatively straightforward to apply this approach for aspirin CRC chemoprevention as well. While calculation of bleeding risk is less well established, there are publicly available calculators that combine cardiovascular and bleeding risk for primary aspirin prevention and such decision aids should be explored for aspirin CRC chemoprevention. However, given the recent recommendation reversal by the USPSTF, I am concerned that recommendation and uptake of aspirin CRC chemoprevention will decline substantially.
In order to reduce CRC burden, we should employ everything in our armamentarium including aspirin chemoprevention. Individualized risk assessment for aspirin chemoprevention, as advised by the AGA practice guideline, will enable the right people to benefit while minimizing risks. Future studies should strengthen the evidence base for aspirin CRC chemoprevention and refine risk stratification, including for younger individuals given the rise in early-onset CRC. The optimal approach to aspirin chemoprevention was best summed up by the foremost expert in the field, Dr. Andy Chan, to the New York Times:7 “we need to think about personalizing who we give aspirin to, and move away from a one-size-fits-all solution”.
Dr. Kupfer is associate professor of medicine, director of the Gastrointestinal Cancer Risk and Prevention Clinic, and codirector of the Comprehensive Cancer Risk and Prevention Clinic at the University of Chicago. She reports no relevant conflicts of interest.
References
1. Liang PS et al. Clin Gastroenterol Hepatol. 2021 Jul;19(7):1327-36.
2. United States Preventive Services Task Force. “Aspirin Use to Prevent Cardiovascular Disease: Preventive Medication.” Accessed April 10, 2022.
3. Burn J et al. Lancet. 2020 Jun 13;395(10240):1855-63.
4. McNeil JJ et al. N Engl J Med. 2018 Oct 18;379(16):1519-28.
5. Guo CG et al. JAMA Oncol. 2021 Mar 1;7(3):428-35.
6. Lloyd KE et al. Prev Med. 2022 Jan;154:106872.
7. Rabin RC. “Aspirin Use to Prevent 1st Heart Attack or Stroke Should Be Curtailed, U.S. Panel Says.” New York Times. Oct. 13, 2021. Accessed April 10, 2022.
