Figures

Self-reported cancer diagnosis. Note patients may indicate more than one cancer.

## All reported carrying a pathogenic or likely pathogenic variant.

Self-reported race/ethnicity. Note patients may report more than one race/ethnicity.

Self-reported genes. Note patients may report more than one gene.

Self-reported genes among Stomach cancer patients. Note patients may report more than one gene.

Self-reported gastric lesions. Note patients may report more than one lesion.

Table 1. Overall demographics and health characteristics

Characteristic	N = 1,972¹
Age
Less than 29	53 (2.7%)
30-39	206 (10%)
40-49	445 (23%)
50-59	495 (25%)
60-69	490 (25%)
70 and over	282 (14%)
Missing	1
Sex
Female	1,880 (95%)
Male	92 (4.7%)
Race_Ethnicity
American Indian, Alaska Native, or First Nations	1 (<0.1%)
Asian or Asian American	29 (1.5%)
Black or African-American	13 (0.7%)
Hispanic White	21 (1.1%)
Hispanic, Latino, or Latinx	47 (2.4%)
Mixed race/ethnicity	28 (1.4%)
Non-Hispanic White	1,812 (92%)
Other	21 (1.1%)
Any_Gastric_Lesion	976 (49%)
H_pylori	135 (6.8%)
Fam_Hist_Stomach_Cancer
Don't know	332 (17%)
No	1,139 (58%)
Yes	501 (25%)
Stomach_Cancer	51 (2.6%)
¹ n (%)

Table 2. Risk factors by any gastric lesion

Any grastric lesion includes any of the following:

Helicobacter Pylori (H. pylori)
Gastritis
Gastric Ulcer
Gastro-esophageal reflux (GERD)
Intestinal Metaplasia
Barrett’s Esophagus
Pernicious Anemia

HR genes: ATM, BARD1, BRCA1, BRCA2, BRIP1, CHEK2, PALB2, RAD51C, RAD51D, NBN

Lynch genes: MLH1, MSH2, MSH6, PMS2, EPCAM

Hereditary Diffuse Gastric Cancer (HDGC) genes: CDH1, CTNNA1

Juvenile Polyposis Syndrome genes: SMAD4, BMPR1A

Peutz-Jeghers Syndrome gene: STK11 (also known as LKB1)

Li-Fraumeni Syndrome gene: TP53

Recessive cancer predisposition syndrome genes: MUTYH, NTHL1

All other genes not included: CDKN2A, ENG, NF1, PTEN

Table 3. Overall Germline Mutation Carriers (by H. pylori)

Characteristic	N = 1,972¹
Stomach_Cancer
No	1,921 (97%)
Yes	51 (2.6%)
H_pylori
No	1,837 (93%)
Yes	135 (6.8%)
Anti_acid_ever
No	859 (44%)
Yes	1,113 (56%)
¹ n (%)

Characteristic	N = 1,972¹
HR
No	436 (22%)
Yes	1,536 (78%)
Lynch
No	1,620 (82%)
Yes	352 (18%)
HDGC
No	1,891 (96%)
Yes	81 (4.1%)
FAP
No	1,935 (98%)
Yes	37 (1.9%)
Juvenile_Polyposis_Syndrome
No	1,969 (100%)
Yes	3 (0.2%)
Peutz_Jeghers_Syndrome
No	1,964 (100%)
Yes	8 (0.4%)
Li_Fraumeni_Syndrome
No	1,962 (99%)
Yes	10 (0.5%)
Recessive_Cancer_Predisposition
No	1,951 (99%)
Yes	21 (1.1%)
Other_genes
No	1,972 (100%)
¹ n (%)

Characteristic	Overall N = 1,972¹	No N = 1,837¹	Yes N = 135¹	p-value
HR				0.4²
No	436 (22%)	410 (22%)	26 (19%)
Yes	1,536 (78%)	1,427 (78%)	109 (81%)
Lynch				0.8²
No	1,620 (82%)	1,510 (82%)	110 (81%)
Yes	352 (18%)	327 (18%)	25 (19%)
HDGC				0.5²
No	1,891 (96%)	1,760 (96%)	131 (97%)
Yes	81 (4.1%)	77 (4.2%)	4 (3.0%)
FAP				0.5³
No	1,935 (98%)	1,801 (98%)	134 (99%)
Yes	37 (1.9%)	36 (2.0%)	1 (0.7%)
Juvenile_Polyposis_Syndrome				0.2³
No	1,969 (100%)	1,835 (100%)	134 (99%)
Yes	3 (0.2%)	2 (0.1%)	1 (0.7%)
Peutz_Jeghers_Syndrome				>0.9³
No	1,964 (100%)	1,829 (100%)	135 (100%)
Yes	8 (0.4%)	8 (0.4%)	0 (0%)
Li_Fraumeni_Syndrome				0.026³
No	1,962 (99%)	1,830 (100%)	132 (98%)
Yes	10 (0.5%)	7 (0.4%)	3 (2.2%)
Recessive_Cancer_Predisposition				>0.9³
No	1,951 (99%)	1,817 (99%)	134 (99%)
Yes	21 (1.1%)	20 (1.1%)	1 (0.7%)
Other_genes
No	1,972 (100%)	1,837 (100%)	135 (100%)
¹ n (%)
² Pearson’s Chi-squared test
³ Fisher’s exact test

Table 4. Cancers by family history of gastric cancer

Note: n=332 who responded “Don’t know” are not included in table 4.

Table 5. Risk factors by gastric cancer status

Characteristic	Overall N = 944¹	No N = 893¹	Yes N = 51¹	p-value
Fam_Hist_Stomach_Cancer				<0.001²
Don't know	140 (15%)	134 (15%)	6 (12%)
No	552 (58%)	533 (60%)	19 (37%)
Yes	252 (27%)	226 (25%)	26 (51%)
Any_Gastric_Lesion				0.044²
No	481 (51%)	462 (52%)	19 (37%)
Yes	463 (49%)	431 (48%)	32 (63%)
H_pylori				0.065³
No	890 (94%)	845 (95%)	45 (88%)
Yes	54 (5.7%)	48 (5.4%)	6 (12%)
HR				<0.001²
No	211 (22%)	182 (20%)	29 (57%)
Yes	733 (78%)	711 (80%)	22 (43%)
Lynch				0.7²
No	799 (85%)	755 (85%)	44 (86%)
Yes	145 (15%)	138 (15%)	7 (14%)
HDGC				<0.001³
No	886 (94%)	855 (96%)	31 (61%)
Yes	58 (6.1%)	38 (4.3%)	20 (39%)
Anti_acid_ever				>0.9²
No	393 (42%)	372 (42%)	21 (41%)
Yes	551 (58%)	521 (58%)	30 (59%)
Anti_acid_time
No	393 (42%)	372 (42%)	21 (42%)
Daily	205 (22%)	185 (21%)	20 (40%)
1-4 times weekly	136 (14%)	136 (15%)	0 (0%)
1-4 times monthly	54 (5.7%)	52 (5.8%)	2 (4.0%)
1-4 times in past 6 months	118 (13%)	115 (13%)	3 (6.0%)
Other/As needed	37 (3.9%)	33 (3.7%)	4 (8.0%)
Missing	1	0	1
¹ n (%)
² Pearson’s Chi-squared test
³ Fisher’s exact test

Let’s look at cell counts for (gastic cancer/no cancer), H pylori, and HR

##                              HR  No Yes
## Stomach_ORNO_Cancer H_pylori           
## No                  No          173 672
##                     Yes           9  39
## Yes                 No           27  18
##                     Yes           2   4

Table 6. Risk factors by (gastric cancer or any gastric lesion)

Cases: Gastric cancer, or any gastric lesion

Controls: (Non-gastric cancer and no family history of gastric cancer) or no cancer

Note we are assuming comparability of groups.

Characteristic	Overall N = 1,457¹	0 N = 462¹	1 N = 995¹	p-value
H_pylori				<0.001²
No	1,322 (91%)	462 (100%)	860 (86%)
Yes	135 (9.3%)	0 (0%)	135 (14%)
Anti_acid_ever				<0.001²
No	507 (35%)	275 (60%)	232 (23%)
Yes	950 (65%)	187 (40%)	763 (77%)
HR				<0.001²
No	355 (24%)	77 (17%)	278 (28%)
Yes	1,102 (76%)	385 (83%)	717 (72%)
Lynch				<0.001²
No	1,179 (81%)	400 (87%)	779 (78%)
Yes	278 (19%)	62 (13%)	216 (22%)
HDGC				0.011²
No	1,385 (95%)	449 (97%)	936 (94%)
Yes	72 (4.9%)	13 (2.8%)	59 (5.9%)
FAP				0.6²
No	1,429 (98%)	452 (98%)	977 (98%)
Yes	28 (1.9%)	10 (2.2%)	18 (1.8%)
Juvenile_Polyposis_Syndrome				0.6³
No	1,454 (100%)	462 (100%)	992 (100%)
Yes	3 (0.2%)	0 (0%)	3 (0.3%)
Peutz_Jeghers_Syndrome				0.3³
No	1,453 (100%)	462 (100%)	991 (100%)
Yes	4 (0.3%)	0 (0%)	4 (0.4%)
Li_Fraumeni_Syndrome				0.11³
No	1,450 (100%)	462 (100%)	988 (99%)
Yes	7 (0.5%)	0 (0%)	7 (0.7%)
Recessive_Cancer_Predisposition				0.2³
No	1,443 (99%)	460 (100%)	983 (99%)
Yes	14 (1.0%)	2 (0.4%)	12 (1.2%)
Stomach_Genes				0.050²
No	40 (2.7%)	7 (1.5%)	33 (3.3%)
Yes	1,417 (97%)	455 (98%)	962 (97%)
Other_genes
No	1,457 (100%)	462 (100%)	995 (100%)
¹ n (%)
² Pearson’s Chi-squared test
³ Fisher’s exact test

Table 7. Associations between cancer-predisposing gene groups and risk for (gastric cancer or any gastric lesion).

Models are adjusted for sex and current age category. 95% CI uses the profile likelihood method.

Gene	No cancer/Non-gastric cancer & No family history n=462	Gastric/Lesion n=995	OR (95% CI)	P-Value
HRYes	385 (83.33)	717 (72.06)	0.42 (0.31-0.57)	0.0000000
LynchYes	62 (13.42)	216 (21.71)	1.94 (1.41-2.69)	0.0000549
HDGCYes	13 (2.81)	59 (5.93)	2.65 (1.46-5.17)	0.0024257
FAPYes	10 (2.16)	18 (1.81)	1.13 (0.51-2.64)	0.7625385
Recessive_Cancer_PredispositionYes	2 (0.43)	12 (1.21)	2.58 (0.66-17.15)	0.2311831
Stomach_GenesYes	455 (98.48)	962 (96.68)	0.32 (0.13-0.71)	0.0090270
BRCA1Yes	100 (21.65)	162 (16.28)	0.67 (0.5-0.9)	0.0080338
BRCA2Yes	161 (34.85)	228 (22.91)	0.53 (0.41-0.68)	0.0000009
CDH1Yes	13 (2.81)	56 (5.63)	2.5 (1.37-4.89)	0.0046260
H_pyloriYes	0 (0)	135 (13.57)	19963182.28 (1.85953575363393e+22-1.283383537954e+64)	0.9590525

Looking at gastric risk scores

Interaction effects - (Unable to do, 0 cell)

Parameter	Odds Ratio	SE	95% CI	z	p
HR (Yes)	0.40	0.06	(0.29, 0.54)	-5.98	< .001
H pylori (Yes)	1.12e+07	8.35e+09	(0.00, Inf)	0.02	0.983
HR (Yes) × H pylori (Yes)	2.14	1776.37	(0.00, Inf)	9.20e-04	> .999

Table 8. Stomach Cancer: Interaction effects of HR and HDGC genes with H. pylori any gastric lesion

Models adjusted for sex and entry age category. 95% CI uses the profile likelihood method. Note that interaction analysis is likely underpowered due to low n=51 stomach cases.

8a. HR x H. pylori

##              Stomach_ORNO_Cancer  No Yes
## HR  H_pylori                            
## No  No                           173  27
##     Yes                            9   2
## Yes No                           672  18
##     Yes                           39   4

Parameter	Odds Ratio	SE	95% CI	z	p
HR (Yes)	0.15	0.05	(0.08, 0.29)	-5.72	< .001
H pylori (Yes)	1.50	1.26	(0.21, 6.65)	0.48	0.631
HR (Yes) × H pylori (Yes)	2.42	2.48	(0.34, 22.09)	0.86	0.389

8b. HR x any gastric lesion

##                        Stomach_ORNO_Cancer  No Yes
## HR  Any_Gastric_Lesion                            
## No  No                                      77  13
##     Yes                                    105  16
## Yes No                                     385   6
##     Yes                                    326  16

Parameter	Odds Ratio	SE	95% CI	z	p
HR (Yes)	0.08	0.04	(0.03, 0.23)	-4.73	< .001
Any Gastric Lesion (Yes)	0.86	0.36	(0.38, 1.96)	-0.37	0.710
HR (Yes) × Any Gastric Lesion (Yes)	3.33	2.14	(0.97, 12.28)	1.87	0.061

8c. HDGC x H. pylori

##               Stomach_ORNO_Cancer  No Yes
## HDGC H_pylori                            
## No   No                           807  25
##      Yes                           48   6
## Yes  No                            38  20
##      Yes                            0   0

Parameter	Odds Ratio	SE	95% CI	z	p
HDGC (Yes)	18.69	6.94	(9.04, 39.05)	7.89	< .001
H pylori (Yes)	3.98	1.94	(1.40, 9.78)	2.83	0.005

8d. HDGC x any gastric lesion

##                         Stomach_ORNO_Cancer  No Yes
## HDGC Any_Gastric_Lesion                            
## No   No                                     449   6
##      Yes                                    406  25
## Yes  No                                      13  13
##      Yes                                     25   7

Parameter	Odds Ratio	SE	95% CI	z	p
HDGC (Yes)	87.29	53.08	(27.76, 308.66)	7.35	< .001
Any Gastric Lesion (Yes)	4.37	2.02	(1.88, 11.91)	3.19	0.001
HDGC (Yes) × Any Gastric Lesion (Yes)	0.05	0.04	(0.01, 0.24)	-3.71	< .001

Table 9. Colorectal Cancer: Interaction effects of HR and Lynch genes with H. pylori

Models adjusted for sex and entry age category. 95% CI uses the profile likelihood method. Note that interaction analysis is likely underpowered due to low n=88 colorectal cases. There are cells with 0 value.

9a. HR x H. pylori

## , ,  = No
## 
##      
##         No  Yes
##   No   356   22
##   Yes 1404  102
## 
## , ,  = Yes
## 
##      
##         No  Yes
##   No    54    4
##   Yes   23    7

Parameter	Odds Ratio	SE	95% CI	z	p
HR (Yes)	0.09	0.02	(0.05, 0.15)	-9.16	< .001
H pylori (Yes)	1.14	0.67	(0.31, 3.28)	0.22	0.824
HR (Yes) × H pylori (Yes)	3.54	2.62	(0.85, 16.35)	1.71	0.087

9b. Lynch x H. pylori

## , ,  = No
## 
##      
##         No  Yes
##   No  1485  108
##   Yes  275   16
## 
## , ,  = Yes
## 
##      
##         No  Yes
##   No    25    2
##   Yes   52    9

Parameter	Odds Ratio	SE	95% CI	z	p
Lynch (Yes)	12.22	3.17	(7.43, 20.61)	9.66	< .001
H pylori (Yes)	1.02	0.76	(0.16, 3.54)	0.03	0.976
Lynch (Yes) × H pylori (Yes)	3.19	2.81	(0.66, 23.90)	1.32	0.187

##                                                  H_pylori   No  Yes
## Race_Ethnicity                                                     
## American Indian, Alaska Native, or First Nations             1    0
## Asian or Asian American                                     24    5
## Black or African-American                                   13    0
## Hispanic White                                              20    1
## Hispanic, Latino, or Latinx                                 36   11
## Mixed race/ethnicity                                        23    5
## Non-Hispanic White                                        1700  112
## Other                                                       20    1

Locations of the 41 NHWs that have H. pylori

## 
##                         California                           Colorado 
##                                 13                                  3 
##                        Connecticut                            Florida 
##                                  2                                  7 
##                            Georgia                           Illinois 
##                                  1                                  2 
##                               Iowa                          Isla Wake 
##                                  1                                  1 
##                             Kansas                           Kentucky 
##                                  2                                  1 
##                              Maine                           Maryland 
##                                  2                                  2 
##                           Michigan                          Minnesota 
##                                  1                                  2 
##                           Missouri                            Montana 
##                                  2                                  1 
##                      New Hampshire                         New Jersey 
##                                  2                                  4 
##                           New York                     North Carolina 
##                                  4                                 13 
##                       North Dakota                               Ohio 
##                                  1                                  6 
##                             Oregon Outside of the US (please specify) 
##                                  2                                  4 
##                       Pennsylvania                Prefer not to share 
##                                  2                                  9 
##                     South Carolina                              Texas 
##                                  1                                  7 
##                               Utah                           Virginia 
##                                  1                                  2 
##                         Washington                      West Virginia 
##                                  7                                  1 
##                          Wisconsin 
##                                  3

Exploring the outcome and exposures

Considering Dr. Nuno’s supervised learning with penalized logistic regression (LASSO)

## [1] 0.03831503

## 23 x 1 sparse Matrix of class "dgCMatrix"
##                      s1
## (Intercept) -0.09850932
## Q4           .         
## Q5          -1.36965693
## Q6           .         
## Q7           .         
## Q8          -1.30118070
## Q10          .         
## Q12          .         
## Q13         -0.55436438
## Q15          .         
## Q16          .         
## Q17          .         
## Q18         -0.05233716
## Q19         -0.56803125
## Q20          .         
## Q21          .         
## Q22         -1.75884004
## Q23          .         
## Q24          .         
## Q25          0.15992353
## Q26          .         
## Q27          .         
## Q28          0.15992331

## 23 x 1 Matrix of class "dgeMatrix"
##                    s1
## (Intercept) 0.9061872
## Q4          1.0000000
## Q5          0.2541942
## Q6          1.0000000
## Q7          1.0000000
## Q8          0.2722102
## Q10         1.0000000
## Q12         1.0000000
## Q13         0.5744373
## Q15         1.0000000
## Q16         1.0000000
## Q17         1.0000000
## Q18         0.9490088
## Q19         0.5666399
## Q20         1.0000000
## Q21         1.0000000
## Q22         0.1722445
## Q23         1.0000000
## Q24         1.0000000
## Q25         1.1734211
## Q26         1.0000000
## Q27         1.0000000
## Q28         1.1734209

## 0.8304813

## 
## Call:
## glm(formula = gc_outcome ~ ., family = "binomial", data = df2[, 
##     c("gc_outcome", c(selected_vars, "age", "Sex"))])
## 
## Coefficients:
##               Estimate Std. Error z value Pr(>|z|)    
## (Intercept)  -12.31939    2.23359  -5.516 3.48e-08 ***
## Q5No           3.10178    0.81589   3.802 0.000144 ***
## Q8No           2.67589    0.58247   4.594 4.35e-06 ***
## Q13No          2.17984    0.67386   3.235 0.001217 ** 
## Q18No          1.57976    0.78185   2.021 0.043327 *  
## Q19No          2.26627    0.77210   2.935 0.003333 ** 
## Q22PMS2      -18.36253 1129.64401  -0.016 0.987031    
## Q25RAD51D     16.77219 3956.18034   0.004 0.996617    
## Q28TP53       16.72468 3956.18041   0.004 0.996627    
## age            0.02977    0.01851   1.608 0.107856    
## SexMale        0.22611    0.76354   0.296 0.767132    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 201.06  on 160  degrees of freedom
## Residual deviance: 140.86  on 150  degrees of freedom
## AIC: 162.86
## 
## Number of Fisher Scoring iterations: 16

##  (Intercept)         Q5No         Q8No        Q13No        Q18No        Q19No 
## 4.464346e-06 2.223754e+01 1.452532e+01 8.844923e+00 4.853780e+00 9.643343e+00 
##      Q22PMS2    Q25RAD51D      Q28TP53          age      SexMale 
## 1.059872e-08 1.923398e+07 1.834160e+07 1.030214e+00 1.253708e+00

Characteristic	OR	95% CI	p-value
Q5
ATM	—	—
No	22.2	5.41, 154	<0.001
Q8
BRCA1	—	—
No	14.5	4.97, 50.3	<0.001
Q13
CHEK2	—	—
No	8.84	2.55, 37.4	0.001
Q18
MSH2	—	—
No	4.85	1.12, 26.1	0.043
Q19
MSH6	—	—
No	9.64	2.38, 52.8	0.003
Q22
No	—	—
PMS2	0.00		>0.9
Q25
No	—	—
RAD51D	19,233,981	0.00,	>0.9
Q28
No	—	—
TP53	18,341,603	0.00,	>0.9
age	1.03	0.99, 1.07	0.11
Sex
Female	—	—
Male	1.25	0.28, 5.83	0.8
Abbreviations: CI = Confidence Interval, OR = Odds Ratio

Unsupervised

##         y
## clusters  0  1
##        1 79 46
##        2 31  5

Cancer cases do not cleanly separate and are scattered in multiple regions. Some mild grouping but no distinct clusters. Mutation profiles do not strongly separate gastric cancer vs non-cancer patients in a linear low-dimensional structure (PCA). Clustering structure exists, but is weak and not cleanly distinct in PCA space

Principal component analysis of mutation profiles did not reveal clear separation between gastric cancer and non-cancer patients. Similarly, hierarchical clustering showed only weak group structure, suggesting that gastric cancer risk is not driven by global mutation patterns but rather by specific genetic alterations

PCA to detect global variance structure is weak/none. Clustering to see patient subgroups is weak. LASSO to detect sparse predictive features has some strong signals.

Our data suggests that there are no clear groups, but there are a few important genes. This pattern is common in biomedical data as diseases are often not cluster-driven, but are feature-driven for specific mutations.

FORCE Analysis

LCC Lab Community Team & April Vang

2026-08-03