Abstract
BACKGROUND & AIMS:
Lynch syndrome (LS) is associated with increased risks of various gastrointestinal, gynecologic, genitourinary, and other cancers. Many clinical practice guidelines recommend that LS carriers’ screening strategies be devised based on their family history of various cancers, in addition to age-, sex-, and gene-specific considerations. The aim of this study was to examine the association between family history and other clinical factors with LS carriers’ histories of various cancers.
METHODS:
Two cohorts of LS carriers were analyzed: a laboratory-based cohort of consecutively ascertained individuals undergoing germline LS testing and a clinic-based cohort of LS carriers undergoing clinical care at an academic medical center. Multivariable logistic regression was performed to assess clinical factors associated with LS carriers’ histories of various cancers/neoplasms. Familial burden was defined as LS carriers’ aggregate number of first-/second-degree relatives with a history of a given malignancy.
RESULTS:
Multivariable analysis of the laboratory-based cohort (3828 LS carriers) identified familial burden as being incrementally associated with LS carriers’ personal history of endometrial (odds ratio [OR], 1.37 per affected first-/second-degree relative; 95% confidence interval [CI], 1.21–1.56), urinary tract (OR, 2.72; 95% CI, 2.02–3.67), small bowel (OR, 3.17; 95% CI, 1.65–6.12), gastric (OR, 1.93; 95% CI, 1.24–3.02), and pancreaticobiliary cancers (OR, 2.10; 95% CI, 1.21–3.65) and sebaceous neoplasms (OR, 7.39; 95% CI, 2.71–20.15). Multivariable analysis of the clinic-based cohort (607 LS carriers) confirmed a significant association of familial burden of endometrial and urinary tract cancers.
CONCLUSIONS:
Familial burden – in addition to age, sex, and specific LS gene – should be used to assess LS carriers’ risks of specific cancers and guide decision-making about organ-specific surveillance.
Keywords: HNPCC, Extracolonic, Screening
Lynch syndrome (LS) is one of the most common hereditary cancer predisposition syndromes, with an estimated population prevalence of 1:279.1 LS is defined by the presence of a pathogenic germline variant (PGV) in one of the DNA mismatch repair (MMR) genes (MLH1, MSH2, MSH6, PMS2) or EPCAM, leading to an increased lifetime risk of multiple cancer types, including those of the gastrointestinal (colorectal, gastric, small bowel, pancreaticobiliary cancers), gynecologic (endometrial and ovarian cancer) and urinary (ureter, renal pelvis, bladder, and possibly kidney cancers) tracts, brain, and cutaneous sebaceous neoplasms.1–4 Although colonoscopic surveilance5 and aspirin chemoprevention6 reduce the incidence of LS-associated colorectal cancer and risk-reducing hysterectomy/salpingo-oophorectomy reduce the incidence of LS-associated endometrial and ovarian cancer,7 optimal surveillance and risk-reduction approaches for other less common LS-associated malignancies remain undefined. These less common LS-associated cancers have inferior prognoses than LS-associated colorectal, endometrial, and ovarian cancers, and this lack of data thus remains a critical problem.4
Due to this paucity of data, some clinical guidelines recommend against surveillance for LS-associated extracolonic, extra-gynecologic malignancies entirely, whereas other guidelines recommend that family history of specific cancers be used to identify LS carriers for organ-specific cancer surveillance.8–11 It remains largely unknown, however, whether family history is a useful tool for risk stratifying LS carriers’ risks of such cancers. For reasons that remain poorly understood, certain families with LS appear to be disproportionately impacted by specific malignancies. This prompted us to evaluate whether familial burden – the number of individuals in a given family impacted by a specific malignancy, rather than simply the presence or absence of any family history of a specific cancer – is a potential predictor for LS carriers’ risks of specific cancers. We recently identified an independent and incremental association between familial burden of gastric and urothelial cancers with LS probands’ likelihood of these specific malignancies, independent of age, sex, and underlying MMR gene.12,13 The aim of this study was to examine the respective impacts of sex, MMR gene, age, and familial burden in understanding the risk of a wide spectrum of LS-associated cancers/neoplasms using 2 large cohorts of individuals with LS.
Methods
As previously described,12,13 we identified LS carriers from a cohort of consecutively ascertained individuals who were not known to be related and underwent syndrome-specific (rather than multi-syndrome) clinical germline testing of 2 or more LS genes (MLH1, MSH2, MSH6, PMS2, and EPCAM) at a commercial laboratory (Myriad Genetic Laboratories, Inc., Salt Lake City, UT) between June 2006 and July 2013. This laboratory-based cohort served as the primary study population for this analysis. Clinical data (including age at genetic testing, sex, ethnicity, personal history of cancer, and family history of cancer in first- and second-degree relatives [FDRs and SDRs, respectively])) were obtained from test requisition forms completed by the health-care provider ordering LS germline testing.12,13 Individuals with missing clinical data or with multiple LS gene PGVs were excluded, as previously described.12,13 Germline testing and variant classification methodologies were performed by Myriad Genetic Laboratories, Inc (Salt Lake City, UT).12–14
An additional clinic-based cohort of LS carriers was analyzed to validate the findings from the laboratory-based cohort. This clinic-based cohort consisted of individuals with confirmed PGVs in MLH1, MSH2, MSH6, PMS2, or EPCAM who were seen for clinical LS care at the Dana-Farber Cancer Institute and consented to participation in an institutional cancer genetics registry from January 2000 and March 2020. In families with multiple members on the registry, we assigned proband status to the LS carrier tested at our site with the most comprehensive clinical and/or genetic testing information in the registry’s database. Collected data included age at testing, sex, ethnicity, personal history of cancer, and family history of cancer in first- and second-degree relatives from the affected side of the family, all of which were ascertained as part of routine clinical care. Personal and family history were self-reported, reviewed by the genetic counselor and physician providing care to the index patient, and entered in a standardized manner into a database using the Progeny (Progeny Genetics, Delray Beach, FL) software program.
For both cohorts, subject age was defined as age at time genetic testing was ordered. On multivariable analyses, LS carriers with MSH2 and EPCAM PGVs were analyzed in aggregate as were those with MSH6 and PMS2 PGVs due to sample size. Analyses of LS carriers with personal histories of endometrial and ovarian cancer were limited to females. For both cohorts, cancers of the ureter, renal pelvis, bladder, and kidney were collectively defined as “urinary tract cancers,” cancers of the duodenum, ileum, and jejunum were collectively defined as “small bowel cancers,” cancers of the pancreas and hepatobiliary tract were collectively defined as “pancreaticobiliary cancers,” and sebaceous adenomas and sebaceous carcinomas of the skin were collectively defined as “sebaceous neoplasms.” The following 9 types of cancer/neoplasia were considered “LS-associated cancers,” for the purposes of this study: colorectal, endometrial, ovarian, urinary tract, small bowel, gastric, pancreaticobiliary, and brain cancers, and sebaceous neoplasms. Family history of each cancer type was analyzed as a categorical (yes/no) variable. Familial burden was defined as the number of first- and second-degree relatives with a specific cancer from the same side of the family and was analyzed as a continuous variable.
Descriptive statistics were used to summarize both cohorts. Clinical factors (age, sex, MMR gene, familial burden) possibly associated with LS carriers’ personal histories of 9 specific LS-associated cancers (colorectal, endometrial, ovarian, gastric, urinary tract, small bowel, pancreaticobiliary, and brain cancer and sebaceous neoplasms) were assessed using multivariable logistic regression models via their odds ratios (OR) with 95% confidence intervals (CI). We considered the association to be statistically significant at alpha <.05 (2-sided), when the CI for OR did not include 1. SAS statistical software version 9.4 (SAS Institute, Inc. Cary, NC) was used for data management and data derivation, and R version 4.0.1 (R Foundation for Statistical Computing, Vienna, Austria) was used for conducting descriptive analyses and multivariable logistic regression. The multivariable analysis was performed for each of the 9 LS-associated cancers with the laboratory-based cohort. For the clinic-based cohort, the analysis was limited to colorectal, endometrial, ovarian, and urinary tract cancers and sebaceous neoplasms because there were insufficient cases of other cancer types for reliable estimates of model parameters (threshold was set a priori at ≥25 cases of a given cancer, for the purposes of this analysis). The study was approved by the Dana-Farber/Harvard Cancer Center Institutional Review Board (Boston, MA).
Results
The laboratory-based cohort consisted of 3828 individuals (63.5% females; 59.3% white; median age, 50.0 years) with PGVs in MLH1 (n = 1346), MSH2 (n = 1639), MSH6 (n = 670), PMS2 (n = 145), or EPCAM (n = 28), as previously described (Table 1).12,13 Also, 3186 of 3828 (83.2%) individuals had a personal history of any LS-associated cancer, including colorectal cancer (n = 2496; 65.2%), endometrial cancer (n = 908/2431 female carriers; 37.4%), ovarian cancer (n = 197/2431 female carriers; 8.1%), urinary tract cancer (n = 158; 4.1%), sebaceous neoplasia (n = 143; 3.7%), small bowel cancer (n = 56; 1.5%), gastric cancer (n = 41; 1.1%), pancreaticobiliary cancer (n = 31; 0.8%), and brain cancer (n = 18; 0.5%). Of the 3828 LS probands in this cohort, 2997 (78.3%) had ≥1 FDR/SDR with colorectal cancer, 966 (25.2%) endometrial cancer, 411 (10.7%) ovarian cancer, 369 (9.6%) urinary tract cancer, 19 (0.5%) sebaceous neoplasia, 70 (1.8%) small bowel cancer, 350 (9.1%) gastric cancer, 351 (9.2%) pancreaticobiliary cancer, and 189 (4.9%) brain cancer (Table 1). The distributions of these characteristics by the LS carriers’ personal histories of the 9 specific LS-associated cancers are shown in Supplementary Table 1.
Table 1.
Clinical Characteristics of Laboratory- and Clinic-Based Cohorts of Individuals With LS
| Laboratory-based cohort (N = 3828) | Clinic-based cohort (N = 607) | |
|---|---|---|
| n (%) | n (%) | |
| Female | 2431 (63.5) | 413 (68.0) |
| Male | 1397 (36.5) | 194 (32.0) |
| Median age at germline testing, y (IQR) | 50.0 (41.0, 59.0) | 50.0 (39.5, 61.0) |
| Race | ||
| White | 2271 (59.3) | 501 (82.5) |
| Black/African American | 185 (4.8) | 11 (1.8) |
| Asian | 111 (2.9) | 15 (2.5) |
| Other | 528 (13.8) | 19 (3.1) |
| Missing/no answer | 733 (19.1) | 61 (10.0) |
| MMR gene | ||
| MLH1 | 1346 (35.2) | 137 (22.6) |
| MSH2 | 1639 (42.8) | 180 (29.7) |
| MSH6 | 670 (17.5) | 145 (23.9) |
| PMS2 | 145 (3.8) | 138 (22.7) |
| EPCAM | 28 (0.7) | 7 (1.2) |
| Personal history of cancer | ||
| Any LS-associated cancer | 3186 (83.2) | 371 (61.1) |
| Colorectal cancer | 2496 (65.2) | 211 (34.8) |
| Endometrial cancera | 908 (37.4)a | 129 (31.2)a |
| Ovarian cancera | 197 (8.1)a | 30 (7.3)a |
| Urinary tract cancer | 158 (4.1) | 27 (4.4) |
| Sebaceous neoplasia | 143 (3.7) | 39 (6.4) |
| Small bowel cancer | 56 (1.5) | 10 (1.6) |
| Gastric cancer | 41 (1.1) | 9 (1.5) |
| Pancreaticobiliary cancer | 31 (0.8) | 15 (2.5) |
| Brain cancer | 18 (0.5) | 4 (0.7) |
| Family history of cancerb | ||
| Colorectal cancer | 2997 (78.3) | 405 (66.7) |
| Endometrial cancer | 966 (25.2) | 219 (36.1) |
| Ovarian cancer | 411 (10.7) | 72 (11.9) |
| Urinary tract cancer | 369 (9.6) | 126 (20.8) |
| Sebaceous neoplasia | 19 (0.5) | 33 (5.4) |
| Small bowel cancer | 70 (1.8) | 21 (3.5) |
| Gastric cancer | 350 (9.1) | 69 (11.4) |
| Pancreaticobiliary cancer | 351 (9.2) | 88 (14.5) |
| Brain cancer | 189 (4.9) | 15 (2.5) |
IQR, interquartile range.
Denominator for female Lynch carriers only.
Family history in any first- and/or second-degree relative.
The clinic-based cohort consisted of 607 individuals (68.0% female; 82.5% white; median age, 50.0 years) in PGVs in MLH1 (n = 137), MSH2 (n = 180), MSH6 (n = 145), PMS2 (n = 138), or EPCAM (n = 7). Also, 371/607 (61.1%) individuals had a personal history of any LS-associated cancer, including colorectal cancer (n = 211; 34.8%), endometrial cancer (n = 129/413 female carriers; 31.2%), ovarian cancer (n = 30/413 female carriers; 7.3%), urinary tract cancer (n = 27; 4.4%), sebaceous neoplasia (n = 39; 6.4%), small bowel cancer (n = 10; 1.6%), gastric cancer (n = 9; 1.5%), pancreaticobiliary cancer (n = 15; 2.5%), and brain cancer (n = 4; 0.7%). Of the 607 LS probands in this cohort, 405 (66.7%) had ≥1 FDR/SDR with colorectal cancer, 219 (36.1%) endometrial cancer, 72 (11.9%) ovarian cancer, 126 (20.8%) urinary tract cancer, 33 (5.4%) sebaceous neoplasia, 21 (3.5%) small bowel cancer, 69 (11.4%) gastric cancer, 88 (14.5%) pancreaticobiliary cancer, and 15 (2.5%) brain cancer (Table 1). The distributions of these characteristics by the LS carriers’ personal histories of the 9 specific LS-associated cancers are shown in Supplementary Table 2.
On multivariable analysis of the laboratory-based cohort, male sex was significantly associated with LS carriers’ likelihood of having colorectal (OR, 4.82; 95% CI, 4.06–5.72), gastric (OR, 2.96; 95% CI, 1.55–5.64), urinary tract (OR, 1.93; 95% CI, 1.37–2.73), and small bowel cancers (OR, 3.84; 95% CI, 2.17–6.77), and sebaceous neoplasms (OR, 1.83; 95% CI, 1.30–2.58), but not pancreaticobiliary or brain cancers (Table 2). MLH1 PGVs were significantly associated with a personal history of colorectal (OR, 3.46; 95% CI, 2.81–4.27) and gastric cancer (OR, 6.56; 95% CI, 1.51–28.54) and inversely associated with personal history of endometrial cancer (OR, 0.57; 95% CI, 0.45–0.72). MSH2/EPCAM PGVs were significantly associated with personal history of colorectal (OR, 1.77; 95% CI, 1.46–2.13), gastric (OR, 5.22; 95% CI, 1.20–22.65), urinary tract (OR, 3.98; 95% CI, 2.37–6.68), and small bowel cancers (OR, 2.53; 95% CI, 1.03–6.18), and sebaceous neoplasms (OR, 4.33; 95% CI, 2.44–7.67).
Table 2.
Multivariable Analysis of Clinical Factors Associated With Various LS-Associated Cancers in the Laboratory-Based Cohort
| Age (per decade) | Male sex (ref: Female) | MLH1 PGV (ref: MSH6/PMS2 PGV) | MSH2/EPCAM PGV (ref: MSH6/PMS2 PGV) | Familial burden of same cancer (per FDR/SDR with same cancer) | |
|---|---|---|---|---|---|
| OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | |
| Personal history of | |||||
| Colorectal cancer | 1.38 (1.30, 1.47) | 4.82 (4.06, 5.72) | 3.46 (2.81, 4.27) | 1.77 (1.46, 2.13) | 0.81 (0.77, 0.86) |
| Endometrial cancera | 1.66 (1.54, 1.78) | NA | 0.57 (0.45, 0.72) | 0.93 (0.75, 1.15) | 1.37 (1.21, 1.56) |
| Ovarian cancera | 1.17 (1.04, 1.31) | NA | 0.71 (0.46, 1.09) | 1.34 (0.94, 1.93) | 1.32 (1.00, 1.76) |
| Urinary tract cancer | 2.43 (2.11, 2.79) | 1.93 (1.37, 2.73) | 1.16 (0.62, 2.15) | 3.98 (2.37, 6.68) | 2.72 (2.02, 3.67) |
| Small bowel cancer | 1.82 (1.48, 2.22) | 3.84 (2.17, 6.77) | 2.24 (0.88, 5.66) | 2.53 (1.03, 6.18) | 3.17 (1.65, 6.12) |
| Sebaceous neoplasms | 1.63 (1.42, 1.85) | 1.83 (1.30, 2.58) | 1.29 (0.66, 2.52) | 4.33 (2.44, 7.67) | 7.39 (2.71, 20.15) |
| Gastric cancer | 2.08 (1.64, 2.65) | 2.96 (1.55, 5.64) | 6.56 (1.51, 28.54) | 5.22 (1.20, 22.65) | 1.93 (1.24, 3.02) |
| Pancreaticobiliary cancer | 1.59 (1.22, 2.08) | 0.73 (0.33, 1.60) | 2.10 (0.66, 6.70) | 2.32 (0.77, 7.01) | 2.10 (1.21, 3.65) |
| Brain cancer | 1.21 (0.85, 1.73) | 1.99 (0.78, 5.07) | 0.12 (0.01, 1.04) | 1.16 (0.40, 3.32) | 1.56 (0.50, 4.87) |
NA, not applicable; ref, reference.
Analysis was restricted to female Lynch syndrome carriers.
On multivariable analysis in the laboratory-based cohort, familial burden of a specific extracolonic LS-associated cancer was significantly and incrementally associated with LS carriers’ likelihood of having that same cancer for gastric cancer (OR, 1.93 per FDR/SDR with gastric cancer; 95% CI, 1.24–3.02), endometrial cancer (OR, 1.37 per FDR/SDR with endometrial cancer; 95% CI, 1.21–1.56), urinary tract cancer (OR, 2.72 per FDR/SDR with urinary tract cancer; 95% CI, 2.02–3.67), pancreaticobiliary cancer (OR, 2.10 per FDR/SDR with pancreaticobiliary cancer; 95% CI, 1.21–3.65), small bowel cancer (OR, 3.17 per FDR/SDR with small bowel cancer; 95% CI, 1.65–6.12), and sebaceous neoplasm (OR, 7.39 per FDR/SDR with sebaceous neoplasia; 95% CI, 2.71–20.15). Familial burden of ovarian cancer had a nonsignificant association with LS carriers’ likelihood of having ovarian cancer (OR, 1.32 per FDR/SDR with ovarian cancer; 95% CI, 1.00–1.76) and familial burden of brain cancer was not associated with personal history of brain cancer. Familial burden of colorectal cancer showed an inverse association with LS carriers’ likelihood of having had colorectal cancer (OR, 0.81 per FDR/SDR with colorectal cancer; 95% CI, 0.77–0.86).
Multivariable analysis in the clinic-based cohort was limited to assessment of LS carriers with a personal history of colorectal, endometrial, ovarian, and urinary tract cancers, and sebaceous neoplasms because there were insufficient cases with personal histories of gastric, small bowel, pancreaticobiliary, or brain cancers. Male sex was significantly associated with personal history of colorectal cancer (OR, 3.39; 95% CI, 2.33–4.92) but not urinary tract cancer or sebaceous neoplasms (Table 3). MLH1 PGVs were significantly associated with personal history of colorectal cancer (OR, 3.50; 95% CI, 2.09–5.84) and sebaceous neoplasms (OR, 5.56; 95% CI, 1.75–17.65), whereas MSH2/EPCAM PGVs were significantly associated with personal history of colorectal (OR, 2.53; 95% CI, 1.60–3.99), endometrial (OR, 1.79; 95% 1.03–3.11), and urinary tract cancers (OR, 3.74; 95% CI, 1.37–10.23), and sebaceous neoplasms (OR, 11.98; 95% CI, 4.38–32.74). Familial burden of endometrial cancer (OR, 1.49 per FDR/SDR with endometrial cancer; 95% CI, 1.12–1.98) and urinary tract cancer (OR, 2.03 per FDR/SDR with urinary tract cancer; 95% CI, 1.28–3.22) was significantly associated with LS carriers’ personal history of these same cancers. The association between familial burden of sebaceous neoplasms and personal history of sebaceous neoplasia in LS carriers was not statistically significant (OR, 1.90 per FDR/SDR with sebaceous neoplasia; 95% CI, 0.86–4.21) in this clinic-based cohort but the association was in the same direction as we saw in the laboratory-based cohort.
Table 3.
Multivariable Analysis of Clinical Factors Associated With Various LS-Associated Cancers in the Clinic-Based Cohort
| Age (per decade) | Male sex (ref: Female) | MLH1 PGV (ref: MSH6/PMS2 PGV) | MSH2/EPCAM PGV (ref: MSH6/PMS2 PGV) | Familial burden of same cancer (per FDR/SDR with same cancer) | |
|---|---|---|---|---|---|
| OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | |
| Personal history of | |||||
| Colorectal cancer | 1.19 (1.03, 1.36) | 3.39 (2.33, 4.92) | 3.50 (2.09, 5.84 | 2.53 (1.60, 3.99) | 1.08 (0.96, 1.22) |
| Endometrial cancera | 2.20 (1.78, 2.70) | NA | 1.56 (0.82, 2.95) | 1.79 (1.03, 3.11) | 1.49 (1.12, 1.98) |
| Ovarian cancera | 1.18 (0.89, 1.57) | NA | 1.49 (0.56, 3.97) | 1.27 (0.53, 3.09) | 1.03 (0.40, 2.67) |
| Urinary tract cancer | 1.53 (1.11, 2.10) | 1.85 (0.83, 4.14) | 2.32 (0.67, 8.08) | 3.74 (1.37, 10.23) | 2.03 (1.28, 3.22) |
| Sebaceous neoplasms | 2.05 (1.53, 2.75) | 1.58 (0.79, 3.18) | 5.56 (1.75, 17.65) | 11.98 (4.38, 32.74) | 1.90 (0.86, 4.21) |
Analysis was restricted to female LS carriers.
Discussion
Since the original description of Family G in the early 20th century,15 LS has been linked to multiple different malignancies, although the relative rarity of the non-colorectal/nonendometrial cancers has made it particularly challenging for clinicians to provide carriers with personalized risk assessment and preventive recommendations regarding these less common cancers. In this large analysis of 2 complementary cohorts of more than 4400 LS carriers, we identified familial burden of most LS-associated cancers as being incrementally associated with LS carriers’ personal likelihood of having the same malignancy, independent of age, sex, and MMR gene. We also confirmed many of the gene- and sex-specific cancer associations that have been seen in other reports.2–4,16
Up until recently, clinical practice guidelines10,17–20 endorsed a “one size fits all” approach for prevention and surveillance in LS, typically consisting of firm recommendations for colonoscopic surveillance every 1–2 years starting at age 20–25 to mitigate colorectal cancer risk and hysterectomy/salpingo-oophorectomy at the completion of childbearing for gynecologic cancer risk-reduction, regardless of specific underlying MMR gene or other individualized factors. Longitudinal data from the multinational Prospective Lynch Syndrome Database (PLSD) have been particularly important in demonstrating the complex interactions between gene-, sex-, and age-specific risks of specific cancers for LS carriers.3,4 Such data have led to the evolution of clinical practice guidelines8,9,11,21 that now largely feature age- and gene-specific recommendations for colonoscopic surveillance and risk-reducing gynecologic surgery in LS carriers.
The personalization of surveillance and risk-reduction strategies for these other, less common LS-associated cancers, however, remains difficult due mostly to a paucity of data on how best to screen for such malignancies in the first place. To address this, a common approach8 has been to use family history of specific non-colorectal nongynecologic LS-associated cancers to triage which LS carriers should be considered for specialized surveillance for these less common cancers, although there have been minimal data to date as to whether family history is associated with LS cancer risk, independent of gene, sex, and age. In the laboratory-based cohort of this study, we demonstrated that familial burden is significantly and incrementally associated with LS carriers’ likelihood of having endometrial cancer, gastric cancer, urinary tract cancer, small bowel cancer, sebaceous neoplasms, and pancreaticobiliary cancer – independent of age, sex, and gene – with a nonsignificant association with familial burden of ovarian cancer. The magnitude of association was particularly notable for some of the less common LS-associated cancers, with roughly 2-fold increased likelihood of gastric and pancreaticobiliary cancer for each affected FDR/SDR, roughly 3-fold increased likelihood of urinary tract and small bowel cancer per affected FDR/SDR, and more than 7-fold increased likelihood (albeit with wide CIs) for each FDR/SDR with cutaneous sebaceous neoplasia. These data strongly suggest that familial burden is as important as gene- and sex-specific associations in assessing LS carriers’ individualized risks for these cancers.
In addition to the importance of familial burden, our analyses confirmed various sex- and gene-specific associations identified in prior analyses. Specifically, our data support those from prior studies demonstrating that male LS carriers have higher colorectal cancer risks than females, and that MLH1 and MSH2 PGV carriers have significantly higher colorectal cancer risks than MSH6 and PMS2 PGV carriers.3,4,22 Data from the PLSD and others have also demonstrated that MSH2 PGV carriers have particularly high risks for urinary tract cancers, that MLH1 and MSH2 PGV carriers have particularly elevated gastric cancer risks, and that male LS carriers have a significantly increased risk of upper GI tract cancers compared with females.2,3 Our data support such sex- and gene-specific findings with particularly strong links between male sex and gastric and small bowel cancers, MLH1 PGVs and gastric cancer, and MSH2 PGVs and gastric, urinary tract, and small bowel cancers and sebaceous neoplasms. However, we observed no significant sex- or gene-specific associations for LS-associated pancreaticobiliary cancers or brain cancers.
Although the benefit of screening all-comers with LS for these less common component cancers remains uncertain at best, data from the PLSD4 demonstrate that these “other” LS-associated cancers have substantially worse prognoses than LS-associated colorectal and endometrial cancers. It is unclear whether the lack of proven benefit to extracolonic cancer surveillance in LS is due to inherent limitations in the screening modalities themselves, or if instead the relative rarity of these cancers in unselected LS carriers has made it difficult to study surveillance interventions that confer a meaningful benefit. Some recent data, however, do suggest a promise of downstaging from extracolonic cancer surveillance. A report from the German Consortium for Familial Intestinal Cancer demonstrated that 83% of gastric cancers detected using regular EGD surveillance were stage I cancers, vs only 25% of gastric cancers that were detected based on symptoms and in the absence of esophagogastroduodenoscopy (EGD) surveillance.23 Another single-institution report similarly demonstrated that 80% of screen-detected gastroduodenal cancers in a cohort of LS carriers undergoing EGD-based surveillance were stage I malignancies.24 Although acknowledging the ongoing uncertainties about the benefit of surveillance for these extracolonic LS-associated malignancies in all-comers with LS, we would propose that familial burden of extracolonic LS-associated cancer – in addition to sex, gene, and age – should be used to assess LS carriers’ risks of specific cancers and guide decision-making about targeted organ-specific surveillance.
One key aspect of this study worth highlighting is the focus on aggregate familial burden of specific LS-associated cancers, rather than simply the binary presence or absence of any family history of that malignancy. By examining familial burden of specific cancers, these data help draw attention to the notion that certain families with LS syndrome appear to be disproportionately impacted by specific malignancies for reasons that appear to be independent of gene- and sex-specific factors. We can only hypothesize as to the biologic mechanisms responsible for an association between familial burden of specific LS-associated cancers (independent of MMR gene) and LS carriers’ personal histories of the same malignancy. Certainly, shared environmental, behavioral, and lifestyle factors could account for some of the familial clustering of these malignancies. Another admittedly speculative potential explanation for such familial clustering could be that shared heritable immunologic factors (eg, HLA polymorphisms) could influence risk of certain LS-associated cancers, especially given emerging data on mechanisms of immune escape that appear to underlie the development and progression of some LS-associated malignancies.25
Several findings from our multivariable analyses warrant specific comment. In our laboratory-based cohort, we identified that familial burden of colorectal cancer was inversely associated with personal history of colorectal cancer for LS carriers (OR, 0.81; 95% CI, 0.77–0.86). We speculate that individuals with a family history of colorectal cancer may have been more likely to initiate early and frequent colonoscopic surveillance, even prior to being diagnosed with LS, which may account for this apparent protective effect of familial burden. An inverse association between MLH1 PGVs and personal history of endometrial cancer was also observed in our laboratory-based cohort (OR, 0.57; 95% CI, 0.45–0.72; reference: MSH6/PMS2 PGV carriers), which may be a result of up-take of risk-reducing hysterectomy among MLH1 PGV carriers and/or the relatively high risk of endometrial cancer conferred by MSH6 PGVs (in the reference group), especially because MSH6 PGV carriers were over-represented in the subset of LS carriers with a personal history of endometrial cancer. Additionally, we recognize that our clinic-based cohort was not able to validate all of the familial burden associations identified in the laboratory-based cohort due to a small number of subjects with several of the less common LS-associated cancers and overall weaker family histories of LS-associated cancer in this cohort. Nonetheless, the significant association between familial burden and urinary tract and endometrial cancers (plus a nonsignificant association with sebaceous neoplasms, likely due to underpowering) in the validation clinic-based cohort suggest that the larger laboratory-based cohort’s familial burden findings are indeed real, meaningful associations.
The use of 2 separate cohorts is a strength of this analysis, especially since the LS carriers in the clinic-based cohort allowed for verification of probands’ personal cancer histories as part of standard clinical care. The large number of LS probands in both cohorts allowed for in-depth assessment of factors associated with 9 different primary LS-associated cancer/neoplasm sites, controlling for age, sex, gene, and familial burden of cancer. Other key strengths include the fact that the laboratory-based cohort consisted of consecutively ascertained LS probands from across the United States, and that both cohorts included LS carriers with PGVs in all 5 of the LS-associated genes, which allowed for assessment of a comprehensive spectrum of LS-associated cancer risks.
We also recognize that there are key limitations to this study. We lacked data on chemoprevention use, cancer surveillance, and/or risk-reducing surgery uptake, which may have influenced subjects’ personal histories of various cancers. Although our data suggest that specific subsets of LS carriers may be most likely to benefit from dedicated surveillance of LS-associated cancers, we acknowledge that we cannot address the optimal means by which to perform such screening. Furthermore, this was a cross-sectional rather than a longitudinal analysis, and we were thus unable to prospectively assess the impact of familial burden and other clinical factors on future cancer risk. Additionally, in the laboratory-based cohort study of individuals referred specifically for LS testing (rather than multi-gene panel testing or unselected population-based germline testing), there is potential for ascertainment bias in which LS carriers with stronger personal and/or family histories of cancer may be over-represented. Both cohorts were predominantly white, which may limit the generalizability of these data to LS carriers of other races. Last, the familial burden analyses cannot account for the possibility of recall bias, wherein a LS carrier with a personal history of a specific uncommon LS-associated cancer may be more likely to report a family history of the same cancer. It is worth noting, however, that a recent analysis26 examined the accuracy and completeness of family history data collected from commercial laboratory test request forms (as was done for the laboratory-based cohort in this study) for individuals referred for germline testing and confirmed exceedingly high accuracy rates for the reported cancer histories in probands’ FDRs and SDRs.
In summary, this analysis of 2 large complementary cohorts of more than 4400 LS carriers identified familial burden as being independently and incrementally associated with probands’ likelihood of endometrial, gastric, urinary tract, small bowel, and pancreaticobiliary cancers and sebaceous neoplasms. Given the relative rarity of many of these less common LS-associated cancers as well as limitations inherent to retrospective cohort analyses, large multi-institutional prospective studies will ultimately be needed to definitively develop evidence-based surveillance/prevention approaches for such malignancies. Based on the data from this study, familial burden should be taken into account – along with sex-, age-, and gene-specific risk considerations – in identifying which LS carriers should be considered for specialized surveillance for these less common cancers and in ultimately developing prospective studies that can hopefully address the question of how best to screen for such malignancies.
Supplementary Material
WHAT YOU NEED TO KNOW.
BACKGROUND AND CONTEXT
Individuals with Lynch syndrome are at increased risk of colorectal, endometrial, ovarian, gastric, urinary tract, pancreaticobiliary, small bowel, and brain malignancies as well as sebaceous neoplasms of the skin. Clinical practice guidelines recommend that family history be used to aid in risk-stratifying Lynch syndrome carriers’ risks of specific cancers, although it remains unknown if family history is a significant risk factor, independent of age, sex, and gene.
NEW FINDINGS
The authors found that familial burden (defined as the number of first- and second-degree relatives with a specific cancer type) of various Lynch syndrome–associated cancers was incrementally associated with Lynch syndrome carriers’ personal histories of endometrial, urinary tract, small bowel, gastric, and pancreaticobiliary cancers, as well as sebaceous neoplasms, independent of age, sex, and gene.
LIMITATIONS
The optimal surveillance strategies for less common Lynch syndrome–associated malignancies remain unknown. Furthermore, this was a cross-sectional analysis rather than a longitudinal analysis.
IMPACT
Familial burden – in addition to age, sex, and gene – should be used to assess Lynch syndrome carriers’ risks of specific cancers and guide decision-making about organ-specific surveillance.
Acknowledgments
Leah H. Biller and Miki Horiguchi contributed equally to this manuscript.
Funding
Supported by the National Institutes of Health (National Cancer Institute) R01CA132829 (S.S.), and The Pussycat Foundation Helen Gurley Brown Presidential Initiative (C.U.).
Abbreviations used in this paper:
- CI
confidence interval
- EGD
esophagogastroduodenoscopy
- FDR
first-degree relative
- LS
Lynch syndrome
- MMR
mismatch repair
- OR
odds ratio
- PGV
pathogenic germline variant
- PLSD
Prospective Lynch Syndrome Database
- SDR
second-degree relative.
Footnotes
Conflicts of interest
H.U. serves as a consultant or advisory role for Roche. S.S. has served as a consultant for Myriad Genetics, Inc.. M.B.Y. reports research funding from Janssen Pharmaceuticals, one-time consulting/scientific advisory board fees from Janssen Pharmaceuticals, and payment for peer review services from UpToDate. The authors report no other conflicts of interest.
Supplementary Material
Note: To access the supplementary material accompanying this article, visit the online version of Gastroenterology at www.gastrojournal.org, and at https://doi.org/10.1053/j.gastro.2021.03.039.
CRediT Authorship Contributions
Leah H. Biller, MD (Formal analysis: Equal; Investigation: Equal; Methodology: Equal; Writing – original draft: Lead; Writing – review & editing: Equal)
Miki Horiguchi, PhD (Data curation: Supporting; Formal analysis: Equal; Investigation: Supporting; Methodology: Supporting; Writing – original draft: Supporting; Writing – review & editing: Equal)
Hajime Uno, PhD (Formal analysis: Equal; Investigation: Supporting; Methodology: Equal; Software: Lead; Writing – original draft: Supporting; Writing – review & editing: Supporting)
Chinedu Ukaegbu, MBBS, MPH (Conceptualization: Supporting; Data curation: Lead; Formal analysis: Supporting; Funding acquisition: Supporting; Project administration: Equal; Supervision: Equal; Writing – original draft: Supporting; Writing – review & editing: Supporting)
Sapna Syngal, MD, MPH (Conceptualization: Equal; Data curation: Equal; Formal analysis: Supporting; Funding acquisition: Lead; Project administration: Lead; Writing – original draft: Supporting; Writing – review & editing: Supporting); Matthew B. Yurgelun, MD (Conceptualization: Lead; Data curation: Supporting; Formal analysis: Equal; Methodology: Supporting; Supervision: Equal; Writing – original draft: Equal; Writing – review & editing: Equal)
References
- 1.Win AK, Jenkins MA, Dowty JG, et al. Prevalence and penetrance of major genes and polygenes for colorectal cancer. Cancer Epidemiol Biomarkers Prev 2017; 26:404–412. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Engel C, Loeffler M, Steinke V, et al. Risks of less common cancers in proven mutation carriers with lynch syndrome. J Clin Oncol 2012;30:4409–4415. [DOI] [PubMed] [Google Scholar]
- 3.Dominguez-Valentin M, Sampson JR, Seppälä TT, et al. Cancer risks by gene, age, and gender in 6350 carriers of pathogenic mismatch repair variants: findings from the Prospective Lynch Syndrome Database. Genet Med 2020;22:15–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Møller P, Seppälä T, Bernstein I, et al. Cancer risk and survival in path_MMR carriers by gene and gender up to 75 years of age: a report from the Prospective Lynch Syndrome Database. Gut 2018;67:1306–1316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Järvinen HJ, Aarnio M, Mustonen H, et al. Controlled 15-year trial on screening for colorectal cancer in families with hereditary nonpolyposis colorectal cancer. Gastroenterology 2000;118:829–834. [DOI] [PubMed] [Google Scholar]
- 6.Burn J, Sheth H, Elliott F, et al. Cancer prevention with aspirin in hereditary colorectal cancer (Lynch syndrome), 10-year follow-up and registry-based 20-year data in the CAPP2 study: a double-blind, randomised, placebo-controlled trial. Lancet 2020;395:1855–1863. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Schmeler KM, Lynch HT, Chen LM, et al. Prophylactic surgery to reduce the risk of gynecologic cancers in the Lynch syndrome. N Engl J Med 2006;354:261–269. [DOI] [PubMed] [Google Scholar]
- 8.National Comprehensive Cancer Network. Genetic/Familial High-Risk Assessment: Colorectal (Version 1.2020). [Google Scholar]
- 9.Seppälä TT, Latchford A, Negoi I, et al. European guidelines from the EHTG and ESCP for Lynch syndrome: an updated third edition of the Mallorca guidelines based on gene and gender. Br J Surg 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Stoffel EM, Mangu PB, Gruber SB, et al. Hereditary colorectal cancer syndromes: American Society of Clinical Oncology Clinical Practice Guideline endorsement of the familial risk-colorectal cancer: European Society for Medical Oncology Clinical Practice Guidelines. J Clin Oncol 2015;33:209–217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Stjepanovic N, Moreira L, Carneiro F, et al. Hereditary gastrointestinal cancers: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2019;30:1558–1571. [DOI] [PubMed] [Google Scholar]
- 12.Kim J, Braun D, Ukaegbu C, et al. Clinical factors associated with gastric cancer in individuals with Lynch syndrome. Clin Gastroenterol Hepatol 2020;18:830–837e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Wischhusen JW, Ukaegbu C, Dhingra TG, et al. Clinical factors associated with urinary tract cancer in individuals with Lynch syndrome. Cancer Epidemiol Biomarkers Prev 2020;29:193–199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kastrinos F, Uno H, Ukaegbu C, et al. Development and validation of the PREMM5 model for comprehensive risk assessment of Lynch syndrome. J Clin Oncol 2017; 35:2165–2172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Warthin AS. Heredity with reference to carcinoma: as shown by the study of the cases examined in the pathological laboratory of the University of Michigan, 1895–1913. Arch Intern Med 1913;12:546–555. [Google Scholar]
- 16.ten Broeke SW, van der Klift HM, Tops CMJ, et al. Cancer risks for PMS2-associated Lynch syndrome. J Clin Oncol 2018;36:2961–2968. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Syngal S, Brand RE, Church JM, et al. ACG clinical guideline: genetic testing and management of hereditary gastrointestinal cancer syndromes. Am J Gastroenterol 2015;110:223–262; quiz 263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Giardiello FM, Allen JI, Axilbund JE, et al. Guidelines on genetic evaluation and management of Lynch syndrome: a consensus statement by the US Multi-Society Task Force on colorectal cancer. Gastroenterology2014;147:502–526. [DOI] [PubMed] [Google Scholar]
- 19.Vasen HF, Blanco I, Aktan-Collan K, et al. Revised guidelines for the clinical management of Lynch syndrome (HNPCC): recommendations by a group of European experts. Gut 2013;62:812–823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Balmaña J, Balaguer F, Cervantes A, et al. Familial risk-colorectal cancer: ESMO Clinical Practice Guidelines. Ann Oncol 2013;24(Suppl 6):vi73–vi80. [DOI] [PubMed] [Google Scholar]
- 21.Monahan KJ, Bradshaw N, Dolwani S, et al. Guidelines for the management of hereditary colorectal cancer from the British Society of Gastroenterology (BSG)/Association of Coloproctology of Great Britain and Ireland (ACPGBI)/United Kingdom Cancer Genetics Group (UKCGG). Gut 2020;69:411–444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bonadona V, Bonaiti B, Olschwang S, et al. Cancer risks associated with germline mutations in MLH1, MSH2, and MSH6 genes in Lynch syndrome. JAMA 2011;305:2304–2310. [DOI] [PubMed] [Google Scholar]
- 23.Ladigan-Badura S, Vangala DB, Engel C, et al. Value of upper gastrointestinal endoscopy for gastric cancer surveillance in patients with Lynch syndrome. Int J Cancer 2021;148:106–114. [DOI] [PubMed] [Google Scholar]
- 24.Kumar S, Dudzik CM, Reed M, et al. Upper endoscopic surveillance in Lynch Syndrome detects gastric and duodenal adenocarcinomas. Cancer Prev Res (Phila) 2020;13:1047–1054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Binder H, Hopp L, Schweiger MR, et al. Genomic and transcriptomic heterogeneity of colorectal tumours arising in Lynch syndrome. J Pathol 2017; 243:242–254. [DOI] [PubMed] [Google Scholar]
- 26.LaDuca H, McFarland R, Gutierrez S, et al. Quality of clinician-reported cancer history when ordering genetic testing. JCO Clin Cancer Inform 2018;2:1–11. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.