Article Text
PDF
Abstract
Objectives Our study was designed to assess whether paired normal-tumour testing increased access to targeted therapy, clinical trials and influenced cancer screening recommendations given to patients and their families.
Design Prospective cohort study.
Setting Academic cancer centre in the Pacific Northwest region of the USA.
Participants Patients newly diagnosed between 01 January 2021 and 31 December 2022 with cancers of the oesophagus, gastro-oesophageal junction and stomach (CEGEJS) were included. All other cancer diagnoses such as head and neck, duodenal and lower gastrointestinal tract cancers were excluded.
Intervention Paired germline and tumour genetic test within 90 days of new patient visit.
Primary outcome measures Number of targeted therapies received (or not) when eligible, follow-up treatment data and number of inherited predispositions to cancers identified. No secondary outcome measures.
Results Of 42 patients, 32 (76.2%) were eligible for at least one targeted therapy. 19 patients received immunotherapy, when 16 had a biomarker predicting immunotherapy benefit, and benefit of immunotherapy was unclear for 3. Another 11 did not have this biomarker, and 6 of them received immunotherapy. Six pathogenic variants were identified in four high-risk genes. By 01 January 2024, 18 patients (42.9%) had died of complications of cancer.
Conclusion More than 75% of patients who received tumour testing were eligible for a targeted therapy regardless of their stage at diagnosis, emphasising the need to expand access to testing with staging workup to improve survival outcomes. Six families received personalised screening recommendations, thanks to this study.
- Cancer genetics
- CHEMOTHERAPY
- Gastrointestinal tumours
- Molecular aspects
- Health Services
Data availability statement
Data are available upon reasonable request. Data requests should be submitted to the corresponding author.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
This is a prospective cohort characterising 42 patients newly diagnosed with upper gastrointestinal cancers between 01 January 2021 and 31 December 2022.
Retrospective review of claims from major payors was performed to assess characteristics of prior patients with upper gastrointestinal cancers and frequency of genetics referral in our region.
We offered paired germline and tumour genetic testing and assessed its impact on choice of targeted therapy, access to clinical trials and cancer screening recommendations.
Our study is limited to one large academic cancer centre and to genetic testing that was clinically available in 2021.
Sample size was small, limiting our ability to perform comparative analyses between subgroups.
Introduction
Thousands of patients diagnosed with cancers of the oesophagus, gastro-oesophageal junction and stomach (CEGEJS) face a dire prognosis1–3 every year, impelling us to develop better methods for early diagnosis and treatments.
A subset of CEGEJS exhibits mismatch repair (MMR) or homologous DNA damage repair deficiency (dHRD).4 Treatments targeting deficient DNA-repair damage pathways such as immunotherapy and/or poly (ADP-ribose) polymerase (PARP) inhibitors are associated with better tolerance, fewer long-term side effects and better outcomes than conventional cytotoxic chemotherapy and radiation.5 ,6–12 A recent study in advanced gastric cancer where patients with dHRD were treated with neoadjuvant durvalumab (programmed death Ligand −1 inhibitor (PD-L1)), paclitaxel and olaparib (PARP inhibitor) demonstrated promising results.13 14
The aetiology of CEGEJS is heterogeneous and population-dependent.15 16 Familial CEGEJS case studies suggest a hereditary component for up to 15% of patients.17–20 Drawing from the overall survival benefit gained with PARP inhibitors in germline mutated breast and ovarian cancer, understanding inherited genetic factors in CEGEJS would augment our ability to identify the most appropriate targeted therapy and predict response.21 There are rare genetic predispositions to CEGEJS, including hereditary diffuse gastric cancer syndrome (HDGC), tylosis with oesophageal cancer syndrome (TEC) or chromosome breakage disorders.4 22–26 However, patients with more common hereditary cancer syndromes such as Lynch syndrome27 and hereditary breast and ovarian cancer syndrome (HBOC) have an increased lifetime risk of upper gastrointestinal malignancies.22 28–30 Uncovering HBOC would unlock access to targeted treatment with a PARP inhibitor.13 14 Furthermore, a delay in identifying a hereditary cancer syndrome at the time of a patient’s diagnosis closes a window of opportunity for early detection and prevention of hereditary cancers for at-risk relatives. National treatment guidelines, including the National Comprehensive Cancer Network (NCCN) guidelines,31 did not specify guidance for appropriateness of genetics referral for all CEGEJS diagnoses in 2021, limiting access and insurance coverage of genetic services.
The goal of this project was to report on the clinical utility of paired normal-tumour profiling results in guiding choice of therapy, access to clinical trials and assess the prevalence of hereditary cancer syndromes in patients with CEGEJS. With this project, we reviewed retrospective registry and claims data for patients with CEGEJS diagnosed between 2015 and 2019, and we prospectively followed newly diagnosed patients with CEGEJS after they received paired clinical normal-tumour testing.
Methods
This project included a retrospective review of registry and payor claims and a prospective cohort study of patients newly diagnosed with CEGEJS. For the retrospective review, we collected and analysed deidentified health metrics from the Surveillance, Epidemiology and End Results (SEER) data for the 13 counties of the Puget Sound region (see online supplemental figure S1) and claims data submitted to Centers for Medicare & Medicaid Services, Washington state Medicaid, Premera Blue Cross and Regence Blue Shield and shared with Hutchinson Institute for Cancer Outcomes Research (HICOR) between 2015 and 2019. Retrospective dataset contained demographic and ethnicity information, cancer diagnosis and treatment data, family history, payor, area of deprivation index32 33 and reports of referral to genetics or reimbursement for genetic testing for patients diagnosed with CEGEJS. During the prospective cohort study, we estimated the number of patients newly diagnosed with a CEGEJS diagnosis at Fred Hutch by querying an institutionally generated deidentified dashboard of annual completed appointments (see online supplemental figures S4-S6). 2 weeks prior to the study start date, we met with the Fred Hutch gastrointestinal oncologists at each location to share the protocol, eligibility criteria and how to refer to the study. We sent a departmental update on this study after 1 year of enrolment. Between 01 January 2021 and 31 December 2022, gastrointestinal oncologists referred new patients with CEGEJS for a cancer genetics evaluation and study participation. The visit with genetics included collection of demographic information and ancestry, confirmation of histology, construction of a 3-generation family tree, pretest counseling, review of the purpose of the study, documentation of interest for genetic testing and research participation. Following the genetic visit, patients were contacted by a research coordinator who obtained informed consent to participate. Paired somatic and germline genetic testing was ordered by the genetics team and performed using the clinical genetic tests called OncoPlex and BROCA34 35 developed by the Laboratory Medicine at the University of Washington in Seattle, Washington, USA (see online supplemental figure S7). Post-test genetic counseling visit included result disclosure and recommendations for familial cascade testing if indicated. Patients and family members confirmed to have a hereditary cancer syndrome were offered a referral to a gastrointestinal cancer high-risk programme and enrolment in a long-term surveillance programme. The study team performed periodic chart review and recorded participant demographics, personal risk factors, cancer diagnosis based on histology report, treatment sequence, genetic test results and vital status at follow-up. All histologies were included. We also assessed whether each patient met criteria for genetic testing per the NCCN guidelines available in January 2021.36 37 Testing for microsatellite instability (MSI) was performed with next generation sequencing,38 testing for mismatch MMR repair deficiency with immunohistochemistry (IHC) and testing for Human Epidermal Growth Factor 2 (HER2) overexpression with IHC and Fluorescence In Situ Hybridisation (FISH).39 Testing for PD-L1 in a tumour sample was performed by measuring the ratio of tumour cells expressing PD-L1 over the total number of viable tumour cells and reported under a Combined Positive Score (CPS).40 41
Supplemental material
The study was approved by the Institutional Review Board (IRB) of the University of Washington with IRB number: 11 490. Data were stored in a password-protected REDCap database only accessible to the study team. Our study team performed descriptive data analysis using Excel V.2307 and no complex statistical tests were performed. Authors of this manuscript have no competing interests.
Patient and public involvement
The IRB team of the University of Washington includes unaffiliated community members of the Seattle area. They reviewed the protocol for this study. Genetics results for each patient obtained during the study were shared with them, ample time for review and questions was provided. Results of the study will be shared with patients and their families after publication.
Results
Characteristics of patients newly diagnosed with CEGEJS compared with patients diagnosed between 2015 and 2019 in the Puget Sound
Between 01 January 2021 and 31 December 2022, 58 patients completed an appointment at Fred Hutch for a new diagnosis of CEGEJS, see figure 1. 43 patients were referred to our cancer genetics service, and one patient, who was given a diagnosis of laryngeal cancer extending into the upper oesophagus, was excluded. Median age at diagnosis was 59.5 years (range, 33–81 years) with 21 patients (50.0%) aged 30–59 years; 27 patients (64.3%) were male sex compared with 67.4% in our registry from 2015 to 2019; 29 patients (69.0%) were reported of white or European ancestry and 8 patients of Asian descent (19.0%) compared with 79.3% and 9.0%, respectively, in our registry (see table 1 and online supplemental figures S1 and S2).
Demographics and risk factors for gastro-oesophageal cancer in patients newly diagnosed with CEGEJS
Consort diagram for the study on cancers of the oesophagus, gastro-oesophageal junction and stomach (CEGEJS).
Of these 42 patients, 14 (33.3%) had oesophageal cancer, 21 (50.0%) had gastric cancer and 26 (61.9%) had stage 3 or 4 disease at time of diagnosis compared with 41.9% in our registry (see online supplemental figure S2). 12 patients (28.6%) had a prior Helicobacter pylori infection, and 10 (23.8%) had Barrett’s oesophagus. 13 patients (31.0%) had a previous primary cancer diagnosis with breast cancer being the most common prior cancer. Of the 39 patients (92.9%) who had a family history of cancer, 35 patients (81.0%) met the NCCN guideline for genetic testing for HBOC and/or for Lynch syndrome, and 24 patients would not have received germline testing around the time of CEGEJS diagnosis if not referred to cancer genetics through this study, see online supplemental figure S3. 37 patients had Medicare/Medicaid or Tricare, and 30 had a commercial or another insurance. Area Deprivation Index was collected in our payor claims data but not for our prospective cohort as zip codes were not recorded. It was six or greater for 197 patients (35.4%) when most of the inhabitants of the Puget Sound region have an Area Deprivation Index of 3 or lower, see online supplemental figure S2. All patients in our prospective cohort received treatment compared with 424 of 556 patients (76.3%) who received treatment in our registry (online supplemental figure S2). By 1 January 2024, 18 patients (42.9%) had died of complications of CEGEJS.
Tumour profiling and germline genetic results
Through our study, 39 out of 42 patients received tumour genetic testing, see table 2.
Tumour and germline genetic testing results
Six CEGEJS (14.3%) had microsatellite instability-high (MSI-H) and 28 (66.7%) were microsatellite stable (MSS). Of the 6 CEGEJS with MSI-H, 3 patients had documented hypermethylation of the MLH1 promoter, one had somatic biallelic inactivation of MLH1, one with somatic biallelic inactivation of MSH6 and hypermethylation studies were cancelled at patient death for the last patient. All six had negative germline genetic testing. Six CEGEJS (14.3%) had a high Tumour Mutational Burden (TMB >5), TMB for them was between 9 mutations/Mb and 50 mutations/Mb. All 6 of them had concurrent MSI-H. We had no reported MSI status and TMB for 8 and 9 patients, respectively. Reasons for missing tumour profiling data included insufficient tumour content, lost to follow-up, second opinion at Fred Hutch and patient death. A CPS score was documented for 31 of the 42 CEGEJS (73.8%); 26 tumours had a CPS score >1 and 5 had a CPS score ≤1.
Most common somatic pathogenic variants (PVs) identified were in the gene TP53 (53.1%, n=17) followed by KRAS, GRAS and NRAS grouped together (n=8, 25.0%), HER2 (n=6, 18.8%) and MLH1 promoter hypermethylation (n=5, 15.6%). Interestingly, 3 patients had a somatic PV in PIK3CA. One patient had gastro-oesophageal junction cancer and a PIK3CA c.1634 A>G (p.E545G) along with somatic biallelic inactivation of PTEN and KRAS c.175G>A (p.A59T). Two patients had gastric cancer, one with PIK3CA c.3140A >G (p.H1047R) and one with PIK3CA c.323G>A (p.R108H) and KRAS c.38G>A (p.G13D). Five patients (11.9%) had an amplification of CCND1, one in CCNE1 and one in CCND2. No patients received a KRAS inhibitor such as sotorasib (Lumakras) or a PIK3CA inhibitor such as alpelisib (Piqray). One was prescribed the CDK4/6 inhibitor abemaciclib (Verzenio) that was denied by the insurance. One patient was found to have an incidental PV in the gene CSF3R at variant allele fraction (VAF) of 37% that was suspected but not confirmed germline. CSF3R encodes the receptor for granulocyte-colony stimulating factor (G-CSF), is involved in myeloid cell differentiation, and this variant has been associated with lower CSF3R messenger RNA, receptor and response to G-CSF.42 The patient received 5’florouracil based chemotherapy, required G-CSF when his absolute white count nadired below 0.5 and mounted a normal white blood cell count response.
Of 42 patients, 39 (92.8%) received germline genetic testing and three died prior to providing a sample. Six PVs were identified, 2 patients had PVs in genes associated with autosomal recessive conditions, 4 (9.5%) had one or more variants of uncertain significance (VUS), and 29 (69.0%) had negative results. Four patients had germline alterations in the homologous recombination DNA damage/repair pathway with PV in BRCA2, ATM, BRCA1 and biallelic FANCA. One patient with oesophageal cancer before the age of 50 years had a tumour PV in the gene ERCC2 called c.1972C>T (p.R658C) with loss of heterozygosity. There was no history of Xeroderma pigmentosum. One patient with gastric cancer had a PV in the gene FANCA called c.216_217del (p.L72Ffs*7) and a VUS in the gene FANCI called c.839 A >G was identified at VAF 49% on tumour testing. The finding in FANCI was not confirmed to be germline in origin. One patient with gastric cancer before the age of 50 years and their father with a history of gastric cancer shared the same VUS in the PDGFRA called c.470C>T (p.T157I), gene for which there are no functional assays to help clarify its significance. One patient with gastric oesophageal junction cancer had 3 VUSs, one in CTNNA1 called c.1726A >G (p.T576A) which is at a highly evolutionarily conserved position but with limited population and functional data, one splice site variant in the gene USP7 called c.1839+5G>A and one in the gene FBXW7 called c.1076A >G (p.H359R). The gene FBXW7 is a tumour suppressor gene known to be downregulated in gastric cancers. It is being evaluated as a marker for poor prognosis.43 Of the 3 patients who couldn’t receive paired testing, one patient was diagnosed with metastatic diffuse gastric adenocarcinoma with signet ring cells before the age of 40 years. Their tumour was sent to a tumour-only commercial laboratory, and an in-frame deletion in the gene CDH1 called c.1747_1749del (p.L583del) was identified at 47.8% VAF and classified as a VUS. Given the high suspicion for hereditary diffuse gastric cancer syndrome, multiple attempts were made to follow-up without success.
Treatment and targeted therapies
Most patients received surgery alone or neoadjuvant chemotherapy and radiation before surgery when they were eligible. Molecular tumour profiling unlocked access to at least one adjuvant targeted therapy approved by the US Food and Drug Administration for 32 of the 42 patients (76.2%). Targeted therapy was known to be beneficial for 17 patients (40.5%) and potentially beneficial for 21 patients (50.0%) as efficacy was not established yet in gastro-oesophageal cancer but reported in other cancer types. An example of this was having an FGFR2 amplification or a fusion with the potential benefit of erdafitinib (Balversa). Of the 42 patients, 31 patients (61.3%) had a CPS score documented. 19 of them received adjuvant immunotherapy; the tumours of 16 of the 26 patients (61.5%) had a CPS score >1, and 3 had a CPS score ≤1. 11 CEGEJS did not have a CPS score documented, and 6 patients (54.5%) received immunotherapy anyway. Overall, 24 patients (57.1%) received at least one targeted therapy such as pembrolizumab (Keytruda), nivolumab (Opdivo), trastuzumab (Herceptin) and ramucirumab (Cyramza) as part of their first line treatment. Should they need further therapy, 17 patients (40.5%) would be eligible for future clinical trials with a regimen containing a WEE1 kinase inhibitor given TP53 tumour alterations.
Discussion
In our study, we report on the clinical utility of paired normal-tumour genetic testing when performed for all patients newly diagnosed with CEGEJS. In 2021, the NCCN guideline encouraged screening CEGEJS with multiple biomarker tests for eligibility for targeted therapies as part of the standard of care for patients with an advanced diagnosis.17 22 Biomarker testing included testing for HER2 overexpression to prompt considering treatment with trastuzumab,44 testing for MSI or MMR deficiency and PD-L1 to prompt eligibility for adjuvant immune checkpoint inhibitors45 and testing with next generation sequencing panel, when possible, for eligibility to receive novel tyrosine kinase inhibitors. More than 75% of patients who received testing in our study were eligible for a targeted therapy regardless of their stage at diagnosis. Six patients who received trastuzumab had HER2 overexpression in their tumours. Almost three quarters of CEGEJS cases were submitted for a CPS score. 26 patients had a CPS score >1 and only 16 patients received immunotherapy. For the remaining 9 patients, the benefit of immunotherapy was unknown given absent CPS score or CPS score ≤1. Furthermore, a quarter of our patients were found eligible for a novel targeted therapy based on our paired testing that went beyond what is recommended by the NCCN guidelines. Neither CDK4-CDK6 inhibitors nor PIK3CA inhibitors have approval for CEGEJS today. Our data highlight the importance of improving access and utilisation of normal-tumour genetic testing for every CEGEJS to guide treatment decision making21 and to identify better treatment options in the future.
We identified six germline PVs in high-risk genes that would change patients’ eligibility for clinical trials and screening and early detection for their at-risk relatives. Five additional findings were suspicious but lacked either functional data or further work up (CSF3R, CTNNA1, PDGFRA, FANCI and CDH1). More than 80% of patients in our cohort met the HBOC and/or the Lynch syndrome guideline for germline genetic testing. We expected that more patients with CEGEJS would meet the NCCN guidelines for genetic testing for Lynch syndrome, given it is associated with a stronger risk of upper gastrointestinal malignancy compared with HBOC. Of those meeting criteria, less than a third would have been offered germline genetic testing at CEGEJS diagnosis without this study. Still, the number of genetic tests ordered by oncologists was significantly higher than what was found in our retrospective payor data. Less than 2% of patients with CEGEJS, diagnosed between 2015 and 2019 in the Puget Sound region, had any claims for genetic counseling and/or testing. For those who did, they all met the eligibility criteria based on the documented personal or family history. Receipt of genetic counseling in CEGEJS was likely significantly under-reported in the claims data, given that: (1) many patients with CEGEJS do not need to see a genetic counselor to obtain genetic testing through their oncologist or a research study and (2) genetic counseling is not always billable or billed as a service. Findings from this cohort align with other research, showing that 1 in 6 patients with CEGEJS have an actionable hereditary cancer syndrome.46 As more data highlight the prevalence of inherited cancer predispositions for patients with CEGEJS, the NCCN guidelines have updated their recommendations for germline genetic testing. Adding broader guidance on the appropriateness of germline genetic testing for each organ or listing the high-yield and actionable genes in each cancer type may help increase testing uptake. Point-of-care genetic testing may also accelerate the timely identification of patients and relatives with an actionable hereditary cancer syndrome and guide screening for at-risk relatives when they are in a window of opportunity for risk reduction or early detection.
Lastly, it is difficult to know for sure whether the hereditary genetic testing we provide for CEGEJS today is comprehensive. We assume that all cancers develop mutations in the same DNA repair, growth factors and cell cycle pathways. It is possible, however, that inherited alterations in pathways that repair damage caused by alcohol or immunodeficiency that prevent healing from chronic inflammation play a role in carcinogenesis for CEGEJS. The BROCA panel test, for example, did not cover the gene RHBDF2 known to cause autosomal dominant tylosis with oesophageal cancer syndrome, making even this expert test an incomplete genetic evaluation for CEGEJS. Gain-of-function PVs in RHBDF2 are associated with sustained EGFR signalling and dysregulated wound healing in the epidermis and non-keratinised epithelium of the upper gastrointestinal tract.47 48 No patients in our study presented with characteristic features of palmoplantar keratoderma, oral lesions or recurring oesophageal strictures, lowering the probability of detection and causing us to miss this extremely rare diagnosis. Understanding interactions between genetic predispositions affecting chronic healing or repair from environmental exposures would bring powerful insights for cancer treatment and early detection in the future.
Limitations of our project include studying a small sample at one large cancer centre, a short study period during the COVID-19 pandemic, many patients being of white or European ancestry and our claims and SEER data including 13 but not all 39 counties of the state of Washington. It is possible we would have identified additional genetic, personal or environmental risk factors if the study was performed in a broader group of patients of Chinese or Japanese ancestry. Further studies are also needed to understand novel monogenic causes vs polygenic risk markers for CEGEJS, along with the interaction between genetic factors and environmental exposures that increase the risk of developing CEGEJS. A subset of patients with a new CEGEJS eligible for the study was not offered participation. Reasons for why 16 patients were not referred to our study are unknown. We hypothesise that they were not included because: they were diagnosed before 01 January 2021 and came for follow-up care without updated diagnosis codes (from diagnosis of cancer to history of cancer); they had a second opinion but did not establish care; they declined a referral or died before being scheduled; they had testing already or the biopsy was sent to another laboratory for tumour testing among other reasons. We noticed that patients with CEGEJS were referred more often by our main campus oncologists (88.1%, n=37) compared with our community oncologists (11.9%, n=5). Finally, many patients came to the clinic with advanced stage, poor nutritional status and many died before being able to complete their genetic test. Having the ability to store a patient’s DNA in a Clinical Laboratory Improvement Amendments-certified biobank for the future would permit completion of clinical hereditary testing later for the benefit of at-risk relatives.
Conclusion
Our study highlights the yield and downstream impact of paired normal-tumour genetic testing in patients with CEGEJS. Identifying biomarkers unlocked targeted therapeutic options for most of our patients and we hope they will derive improved survival outcomes from these therapies. Uncovering a hereditary cancer syndrome in patients with CEGEJS also allowed for cascade testing, tailored screening, risk reduction and early detection for a broad range of cancers for family members. Further research is needed in the stratification of the risk to develop CEGEJS, genetic modifiers of risk, response to targeted therapy and novel blood-based disease recurrence surveillance tools.
Data availability statement
Data are available upon reasonable request. Data requests should be submitted to the corresponding author.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and the study was approved by the IRB of the University of Washington with IRB number: 11490. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
We want to thank all the patients and families who participated in this study. We also want to thank Catherine Fedorenko and Winona Wright from HICOR and Lauren Santos, MSIM, for helping coordinate the implementation of this project in 2021.
References
Footnotes
X @DrDubardGault
Contributors KT, SC, MYL and MED-G designed the concept for study. LVN, MYL and MED-G prepared our site for this project. SC, LVN, CLH, VS, BO, WMG, BS, EL, LF and MED-G helped with patient care and consent. QS performed the primary analysis from the SEER data for the Puget Sound region. AJ, EQK and CP analysed genetic test results and AJ, EQK, CP and MED-G reviewed and interpreted all genetic results. KT and MED-G completed data analysis. All authors participated in developing the manuscript; all edited and approved the final version of the manuscript. MED-G is the guarantor for this study.
Funding This work was funded by the Brotman Baty Institute for Precision Medicine.
Map disclaimer The depiction of boundaries on this map does not imply the expression of any opinion whatsoever on the part of BMJ (or any member of its group) concerning the legal status of any country, territory, jurisdiction or area or of its authorities. This map is provided without any warranty of any kind, either express or implied.
Competing interests None declared.
Patient and public involvement Patients and/or the public were involved in the design, conduct, reporting or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.