The potential of using artificial intelligence to improve skin cancer diagnoses in Hawai'i's multiethnic population.

Skin cancer remains the most commonly diagnosed cancer in the USA with more than 1 million new cases each year. Melanomas account for about 1% of all skin cancers and most skin cancer deaths. Multiethnic individuals whose skin is pigmented underestimate their risk for skin cancers and melanomas and may delay seeking a diagnosis. The use of artificial intelligence may help improve the diagnostic precision of dermatologists/physicians to identify malignant lesions. To validate our artificial intelligence's efficiency in distinguishing between images, we utilized 50 images obtained from our International Skin Imaging Collaboration dataset (n = 25) and pathologically confirmed lesions (n = 25). We compared the ability of our artificial intelligence to visually diagnose these 50 skin cancer lesions with a panel of three dermatologists. The artificial intelligence model better differentiated between melanoma vs. nonmelanoma with an area under the curve of 0.948. The three-panel member dermatologists correctly diagnosed a similar number of images (n = 35) as the artificial intelligence program (n = 34). Fleiss' kappa (ĸ) score for the raters and artificial intelligence indicated fair (0.247) agreement. However, the combined result of the dermatologists panel with the artificial intelligence assessments correctly identified 100% of the images from the test data set. Our artificial intelligence platform was able to utilize visual images to discriminate melanoma from nonmelanoma, using de-identified images. The combined results of the artificial intelligence with those of the dermatologists support the use of artificial intelligence as an efficient lesion assessment strategy to reduce time and expense in diagnoses to reduce delays in treatment.

Adapting a Skin Cancer Prevention Intervention for Multiethnic Adolescents.

OBJECTIVE: We tested the effectiveness of a school-based skin cancer prevention intervention entitled "SunSafe in the Middle School Years" adapted for multiethnic high school students. METHODS: In Hawai'i, 208 10th graders (51.6% Asian, 30.4% Native Hawaiian/Pacific Islander, 8.4% white, 3.5% Hispanic, 2.7% black) participated. Changes in sun protection knowledge, attitudes, and self-reported behaviors were measured using a standardized 18-item survey. The Systematic Observation of Sun Protection Factors (SOSPF) instrument assessed aggregate sun protection behaviors. RESULTS: At posttest, improvements were found in 13 of 18 survey items (p < .05), and retained in 10 items at 12-months following baseline assessments; sun-protection attitudes and intended tanning behavior did not show improvement. Six observers using SOSPF reliably measured students' sun protection behaviors at school including use of hats, sunglasses, long sleeves, lower body coverage, and shade (ICC > .77). CONCLUSIONS: We uncovered a lack of knowledge about UVR exposure, tanning, and lifetime skin cancer risk among multiethnic high school students. We found that students' tanning attitudes may be influenced by self-perceptions regarding their own complexion, but were willing to modify their sun protection behaviors once informed about skin cancer risk.

Novel colon cancer susceptibility variants identified from a genome-wide association study in African Americans.

Genome-wide association studies (GWAS) in ethnic/racial minority populations can help to fine-map previously identified risk regions or discover new risk loci because of the genetic diversity in these populations. We conducted a GWAS of colorectal cancer (CRC) in 6,597 African Americans (1,894 cases and 4,703 controls) (Stage 1) and followed up the most promising markers in a replication set of 2,041 participants of African descent (891 cases and 1,150 controls) (Stage 2). We identified a novel variant, rs56848936 in the gene SYMPK at 19q13.3, associated with colon cancer risk (odds ratio 0.61 for the risk allele G, p = 2.4 × 10-8 ). The frequency of the G allele was 0.06 in African Americans, compared to <0.01 in Europeans, Asians and Amerindians in the 1000 Genomes project. In addition, a variant previously identified through fine-mapping in this GWAS in the region 19q13.1, rs7252505, was confirmed to be more strongly associated with CRC in the African American replication set than the variant originally reported in Europeans (rs10411210). The association between rs7252505 and CRC was of borderline significance (p = 0.05) in a Hispanic population GWAS with 1,611 CRC cases and 4,330 controls. With the three datasets combined, the odds ratio was 0.84 for the risk allele A (95% confidence interval 0.79-0.89, p = 3.7 × 10-8 ). This study further highlights the importance of conducting GWAS studies in diverse ancestry populations.

Risk of extracolonic cancers for people with biallelic and monoallelic mutations in MUTYH.

Germline mutations in the DNA base excision repair gene MUTYH are known to increase a carrier's risk of colorectal cancer. However, the risks of other (extracolonic) cancers for MUTYH mutation carriers are not well defined. We identified 266 probands (91% Caucasians) with a MUTYH mutation (41 biallelic and 225 monoallelic) from the Colon Cancer Family Registry. Mutation status, sex, age and histories of cancer from their 1,903 first- and 3,255 second-degree relatives were analyzed using modified segregation analysis conditioned on the ascertainment criteria. Compared with incidences for the general population, hazard ratios (HRs) (95% confidence intervals [CIs]) for biallelic MUTYH mutation carriers were: urinary bladder cancer 19 (3.7-97) and ovarian cancer 17 (2.4-115). The HRs (95% CI) for monoallelic MUTYH mutation carriers were: gastric cancer 9.3 (6.7-13); hepatobiliary cancer 4.5 (2.7-7.5); endometrial cancer 2.1 (1.1-3.9) and breast cancer 1.4 (1.0-2.0). There was no evidence for an increased risk of cancers at the other sites examined (brain, pancreas, kidney or prostate). Based on the USA population incidences, the estimated cumulative risks (95% CI) to age 70 years for biallelic mutation carriers were: bladder cancer 25% (5-77%) for males and 8% (2-33%) for females and ovarian cancer 14% (2-65%). The cumulative risks (95% CI) for monoallelic mutation carriers were: gastric cancer 5% (4-7%) for males and 2.3% (1.7-3.3%) for females; hepatobiliary cancer 3% (2-5%) for males and 1.4% (0.8-2.3%) for females; endometrial cancer 3% (2%-6%) and breast cancer 11% (8-16%). These unbiased estimates of both relative and absolute risks of extracolonic cancers for people, mostly Caucasians, with MUTYH mutations will be important for their clinical management.

PMS2 monoallelic mutation carriers: the known unknown.

Germ-line mutations in MLH1, MSH2, MSH6, and PMS2 have been shown to cause Lynch syndrome. The penetrance of the cancer and tumor spectrum has been repeatedly studied, and multiple professional societies have proposed clinical management guidelines for affected individuals. Several studies have demonstrated a reduced penetrance for monoallelic carriers of PMS2 mutations compared with the other mismatch repair (MMR) genes, but clinical management guidelines have largely proposed the same screening recommendations for all MMR gene carriers. The authors considered whether enough evidence existed to propose new screening guidelines specific to PMS2 mutation carriers with regard to age at onset and frequency of colonic screening. Published reports of PMS2 germ-line mutations were combined with unpublished cases from the authors' research registries and clinical practices, and a discussion of potential modification of cancer screening guidelines was pursued. A total of 234 monoallelic PMS2 mutation carriers from 170 families were included. Approximately 8% of those with colorectal cancer (CRC) were diagnosed before age 30, and each of these tumors presented on the left side of the colon. As it is currently unknown what causes the early onset of CRC in some families with monoallelic PMS2 germline mutations, the authors recommend against reducing cancer surveillance guidelines in families found having monoallelic PMS2 mutations in spite of the reduced penetrance.Genet Med 18 1, 13-19.

Risk of colorectal cancer for people with a mutation in both a MUTYH and a DNA mismatch repair gene.

The base excision repair protein, MUTYH, functionally interacts with the DNA mismatch repair (MMR) system. As genetic testing moves from testing one gene at a time, to gene panel and whole exome next generation sequencing approaches, understandin g the risk associated with co-existence of germline mutations in these genes will be important for clinical interpretation and management. From the Colon Cancer Family Registry, we identified 10 carriers who had both a MUTYH mutation (6 with c.1187G>A p.(Gly396Asp), 3 with c.821G>A p.(Arg274Gln), and 1 with c.536A>G p.(Tyr179Cys)) and a MMR gene mutation (3 in MLH1, 6 in MSH2, and 1 in PMS2), 375 carriers of a single (monoallelic) MUTYH mutation alone, and 469 carriers of a MMR gene mutation alone. Of the 10 carriers of both gene mutations, 8 were diagnosed with colorectal cancer. Using a weighted cohort analysis, we estimated that risk of colorectal cancer for carriers of both a MUTYH and a MMR gene mutation was substantially higher than that for carriers of a MUTYH mutation alone [hazard ratio (HR) 21.5, 95% confidence interval (CI) 9.19-50.1; p < 0.001], but not different from that for carriers of a MMR gene mutation alone (HR 1.94, 95% CI 0.63-5.99; p = 0.25). Within the limited power of this study, there was no evidence that a monoallelic MUTYH gene mutation confers additional risk of colorectal cancer for carriers of a MMR gene mutation alone. Our finding suggests MUTYH mutation testing in MMR gene mutation carriers is not clinically informative.

PIK3CA Somatic Mutation Status in Relation to Patient and Tumor Factors in Racial/Ethnic Minorities with Colorectal Cancer.

BACKGROUND: Approximately 10% to 20% of colorectal cancers exhibit somatic mutations in the phosphoinositide-3-kinase, catalytic, alpha polypeptide gene (PIK3CA). We evaluated the relationship of PIK3CA mutation status in colorectal cancer with race/ethnicity, colorectal cancer survival, and other patient and tumor factors. METHODS: This study comprised 377 racial/ethnic minorities with incident invasive colorectal cancer, enrolled in the Colon Cancer Family Registry via population-based cancer registries. Tumor specimens were tested for PIK3CA mutations in exon 9 and 20 hotspots, BRAF p.V600E mutations, and DNA mismatch repair (MMR). In logistic regression models, we evaluated the association between PIK3CA mutation status and race/ethnicity, overall, and by mutation site. Using Cox regression, we evaluated the association between PIK3CA mutation status and survival after colorectal cancer diagnosis. RESULTS: PIK3CA mutations were detected in 42 cases (11%), with a similar prevalence across racial/ethnic groups. Individuals with PIK3CA-mutated colorectal cancer were significantly more likely than those with PIK3CA-wildtype disease to have proximal colon cancer, MMR-deficient tumors, and a germline MMR mutation (P ≤ 0.01). There was no evidence for an association between PIK3CA and overall survival (HR, 0.77; 95% confidence interval, 0.43-1.39). CONCLUSIONS: The prevalence of PIK3CA mutation status in colorectal cancer does not differ according to race/ethnicity, but may vary according to other relevant clinicopathologic and etiologic factors, including germline MMR mutation status, tumor MMR status, and tumor site. IMPACT: These findings underscore the importance of PIK3CA mutation status in colorectal cancer epidemiology and provide evidence that the prevalence of such mutations is similar across several racial/ethnic groups.

Trans-ethnic genome-wide association study of colorectal cancer identifies a new susceptibility locus in VTI1A.

The genetic basis of sporadic colorectal cancer (CRC) is not well explained by known risk polymorphisms. Here we perform a meta-analysis of two genome-wide association studies in 2,627 cases and 3,797 controls of Japanese ancestry and 1,894 cases and 4,703 controls of African ancestry, to identify genetic variants that contribute to CRC susceptibility. We replicate genome-wide statistically significant associations (P<5 × 10(-8)) in 16,823 cases and 18,211 controls of European ancestry. This study reveals a new pan-ethnic CRC risk locus at 10q25 (rs12241008, intronic to VTI1A; P=1.4 × 10(-9)), providing additional insight into the aetiology of CRC and highlighting the value of association mapping in diverse populations.

Risk of colorectal cancer for carriers of mutations in MUTYH, with and without a family history of cancer.

We studied 2332 individuals with monoallelic mutations in MUTYH among 9504 relatives of 264 colorectal cancer (CRC) cases with a MUTYH mutation. We estimated CRC risks through 70 years of age of 7.2% for male carriers of monoallelic mutations (95% confidence interval [CI], 4.6%-11.3%) and 5.6% for female carriers of monoallelic mutations (95% CI, 3.6%-8.8%), irrespective of family history. For monoallelic MUTYH mutation carriers with a first-degree relative with CRC diagnosed by 50 years of age who does not have the MUTYH mutation, risks of CRC were 12.5% for men (95% CI, 8.6%-17.7%) and 10% for women (95% CI, 6.7%-14.4%). Risks of CRC for carriers of monoallelic mutations in MUTYH with a first-degree relative with CRC are sufficiently high to warrant more intensive screening than for the general population.

How do researchers manage genetic results in practice? The experience of the multinational Colon Cancer Family Registry.

There is consensus internationally that research participants should be offered the opportunity to receive clinically relevant genetic information identified through research, but there is little empirical peer-reviewed work documenting this process. We report the experience of conducting genetic research with nearly 35,000 participants in the Colon Cancer Family Registry, based in the USA, Canada, Australia, and New Zealand. Investigators from six multinational sites provided information about disclosure protocols, implementation, and uptake of genetic results and made suggestions to inform practice. Across 5 of the 6 registry sites, 1,634 participants in families with mismatch repair or MutYH gene mutations have been offered results. Participant uptake ranged from 56 to 86 %. Researchers faced significant challenges in the effort to return results. We offer suggestions in five key areas: (1) planning for the disclosure process, (2) participant information, (3) autonomy of participants, (4) monitoring scientific progress, and (5) involvement of stakeholders. Despite increasing discussion of the importance of returning incidental findings from genetic research, this paper highlights the considerable diversity, challenges, and costs faced in practice when returning expected findings with established utility and validity. We argue that more work is needed to ensure that genetic results in research are optimally managed.

Genetic variation in the inflammation and innate immunity pathways and colorectal cancer risk.

BACKGROUND: It is widely accepted that chronic inflammation plays a role in the etiology of colorectal cancer. Using a two-stage design, we examined the associations between colorectal cancer and common variation in 37 key genes in the inflammation and innate immunity pathways. METHODS: In the discovery stage, 2,322 discordant sibships (2,535 cases, 3,915 sibling controls) from the Colorectal Cancer Family Registry were genotyped for more than 600 tagSNPs and 99 single-nucleotide polymorphisms (SNP) were selected for further examination based on strength of association. In the second stage, 351 SNPs tagging gene regions covered by the 99 SNPs were tested in 4,783 Multiethnic Cohort subjects (2,153 cases, 2,630 controls). RESULTS: The association between rs9858822 in the PPARG gene and colorectal cancer was statistically significant at the end of the second stage (OR per allele = 1.36, Bonferroni-adjusted P = 0.045), based on the "effective" number of markers in stage II (n = 306). The risk allele C was common (frequency 0.3) in African Americans but rare (frequency < 0.03) in whites, Japanese Americans, Latinos, and Native Hawaiians. No statistically significant heterogeneity of effects across race/ethnicity, body mass index (BMI) levels, regular aspirin use, or pack-years of smoking was detected for this SNP. Suggestive associations were also observed for several SNPs in close vicinity to rs9858822. CONCLUSIONS: Our results provide new evidence of association between PPARG variants and colorectal cancer risk. IMPACT: Further replication in independent samples is warranted.

Fine-mapping of genome-wide association study-identified risk loci for colorectal cancer in African Americans.

Genome-wide association studies of colorectal cancer (CRC) in Europeans and Asians have identified 21 risk susceptibility regions [29 index single-nucleotide polymorphisms (SNPs)]. Characterizing these risk regions in diverse racial groups with different linkage disequilibrium (LD) structure can help localize causal variants. We examined associations between CRC and all 29 index SNPs in 6597 African Americans (1894 cases and 4703 controls). Nine SNPs in eight regions (5q31.1, 6q26-q27, 8q23.3, 8q24.21, 11q13.4, 15q13.3, 18q21.1 and 20p12.3) formally replicated in our data with one-sided P-values <0.05 and the same risk directions as reported previously. We performed fine-mapping of the 21 risk regions (including 250 kb on both sides of the index SNPs) using genotyped and imputed markers at the density of the 1000 Genomes Project to search for additional or more predictive risk markers. Among the SNPs correlated with the index variants, two markers, rs12759486 (or rs7547751, a putative functional variant in perfect LD with it) in 1q41 and rs7252505 in 19q13.1, were more strongly and statistically significantly associated with CRC (P < 0.0006). The average per allele risk was improved using the replicated index variants and the two new markers (odds ratio = 1.14, P = 6.5 × 10(-16)) in African Americans, compared with using all index SNPs (odds ratio = 1.07, P = 3.4 × 10(-10)). The contribution of the two new risk SNPs to CRC heritability was estimated to be 1.5% in African Americans. This study highlights the importance of fine-mapping in diverse populations.

Quality assessment and correlation of microsatellite instability and immunohistochemical markers among population- and clinic-based colorectal tumors results from the Colon Cancer Family Registry.

The detection of defective mismatch repair (MMR), as assessed by the presence of tumor microsatellite instability (MSI) and/or loss of MMR protein expression by IHC, has been useful for risk assessment, prognosis, and prediction of treatment in patients with colorectal cancer. We analyzed tumors for the presence of defective MMR from 5927 Colorectal Cancer Family Registry patients recruited at six international consortium sites. We evaluated the appropriate percentage instability cutoff used to distinguish the three MSI phenotypes [ie, stable (MSS), low instability (MSI-L), and high instability (MSI-H)]; the sensitivity, specificity, and performance characteristics of individual markers; and the concordance between MSI and IHC phenotypes. Guided by the results of the IHC testing, our findings indicate that the distinction between an MSI-H phenotype from a low-instability or MSS phenotype can best be accomplished by using a cutoff of 30% or greater of the markers showing instability. The sensitivity and specificity of the mononucleotide markers were higher than those of the dinucleotide markers. Specifically, BAT26 and BAT25 had the highest sensitivity (94%) and specificity (98%), and the use of mononucleotide markers alone identified 97% of the MSI-H cases correctly. As expected, the presence of MSI-H correlated with an older age of diagnosis, the presence of tumor in the proximal colon, and female sex.

Cancer risks for monoallelic MUTYH mutation carriers with a family history of colorectal cancer.

Cancer risks for a person who has inherited a MUTYH mutation from only one parent (monoallelic mutation carrier) are uncertain. Using the Colon Cancer Family Registry and Newfoundland Familial Colon Cancer Registry, we identified 2,179 first- and second-degree relatives of 144 incident colorectal cancer (CRC) cases who were monoallelic or biallelic mutation carriers ascertained by sampling population complete cancer registries in the United States, Canada and Australia. Using Cox regression weighted to adjust for sampling on family history, we estimated that the country-, age- and sex-specific standardized incidence ratios (SIRs) for monoallelic mutation carriers, compared to the general population, were: 2.04 (95% confidence interval, CI 1.56-2.70; p < 0.001) for CRC, 3.24 (95%CI 2.18-4.98; p < 0.001) for gastric cancer, 3.09 (95%CI 1.07-12.25; p = 0.07) for liver cancer and 2.33 (95%CI 1.18-5.08; p = 0.02) for endometrial cancer. Age-specific cumulative risks to age 70 years, estimated using the SIRs and US population incidences, were: for CRC, 6% (95%CI 5-8%) for men and 4% (95%CI 3-6%) for women; for gastric cancer, 2% (95%CI 1-3%) for men and 0.7% (95%CI 0.5-1%) for women; for liver cancer, 1% (95%CI 0.3-3%) for men and 0.3% (95%CI 0.1-1%) for women and for endometrial cancer, 4% (95%CI 2-8%). There was no evidence of increased risks for cancers of the brain, pancreas, kidney, lung, breast or prostate. Monoallelic MUTYH mutation carriers with a family history of CRC, such as those identified from screening multiple-case CRC families, are at increased risk of colorectal, gastric, endometrial and possibly liver cancers.

Accuracy of colorectal polyp self-reports: findings from the colon cancer family registry.

INTRODUCTION: Colorectal adenomas and other types of polyps are commonly used as end points or risk factors in epidemiologic studies. However, it is not known how accurately patients are able to self-report the presence or absence of adenomas following colonoscopy. METHODS: Participants in the Colon Cancer Family Registry provided self-reports of recent colorectal cancer (CRC) screening activity, and whether or not they had ever been told they had a polyp. Positive and negative predictive values for polyp self-report were calculated by comparing medical records with self-reports from 488 participants. RESULTS: The positive predictive value for self-reported polyp was 80.9%, and the negative predictive value was 85.8%. The predictive values did not differ by age group or sex, but participants with a previous diagnosis of CRC had a lower negative predictive value (76.2%) than participants with no personal history of CRC (89.0%; P = 0.04). CONCLUSIONS: Predictive values for self-reports of polyps are fairly high, but researchers needing accurate polyp data should obtain medical record confirmation. Pursuing medical records on only those participants self-reporting a polyp could result in an underestimation of the polyp prevalence in a study population.