Impact of optional multidisciplinary tumor board meeting on the mortality of patients with gastrointestinal cancer: A retrospective observational study

Abstract Background Multidisciplinary tumor board meetings (MDTs) have shown a positive effect on patient care and play a role in the planning of care. However, there is limited evidence of the association between MDTs and patient mortality and in‐hospital morbidity for mixed cases of gastrointestinal (GI) cancer. Aim To evaluate the influence of optional MDTs on care of patients with cancer to determine potential associations between MDTs and patient mortality and morbidity. Methods and results This was a retrospective observational study at the referral center of King Abdulaziz University Hospital, Jeddah, Kingdom of Saudi Arabia. Among all adult patients diagnosed with GI cancer from January 2017 to June 2019, 130 patients were included. We categorized patients into two groups: 66 in the control group (non‐MDT) and 64 in the MDT group. The main outcome measure was overall mortality, measured by survival analysis. The follow‐up was 100% complete. Four patients in the MDT group and 13 in the non‐MDT group died (P = .04). The median follow‐up duration was 294 days (interquartile range [IQR], 140‐434) in the non‐MDT group compared with 176 days (IQR, 103‐466) in the MDT group (P = .20). There were no differences in intensive care unit or hospital length‐of‐stay or admission rates. The overall mortality at 2 years was 13% (95% confidence interval [CI], 0.06‐0.66) in the MDT group and 38% (95% CI, 0.10‐0.39) in the non‐MDT group (P = .08). The MDT group showed a 72% (adjusted hazard ratio [HR], 0.28; 95% CI, 0.08‐0.90; P = .03) decrease in mortality over time compared with the non‐MDT group. Conclusions MDTs were associated with decreased mortality over time. Thus, MDTs have a positive influence on patient care by improving survival and should be incorporated into care.


| INTRODUCTION
In 2018, the World Health Organization (WHO) reported that one in every six deaths is attributable to cancer and that the incidence of cancer is 18 million worldwide, with an estimated mortality of 9 million. In the same report, the estimated cancer mortality in the Kingdom of Saudi Arabia was reported to be 10 000. 1 As the number of cancer cases increases, many challenges are raised about treatment modalities that suit patients' needs. Moreover, the rapid growth in treatment modalities has made the setting of a treatment plan more complex, and highly specialized personnel are required to provide optimal care to patients with cancer. 2 One approach to treating cancer is personalized medicine, which replaces "one size fits all" models with individualized therapy. 3 Another approach is patient-centered care, which involves the provision of qualified, trustworthy, inter-professional cancer care teams matched with patient needs, beliefs, and priorities, ensuring coordination between patient care teams and caregivers. 2 Multidisciplinary management, which provides the advantage of having physicians from multiple specialties involved in care scheduling, is applied through multidisciplinary teams or multidisciplinary tumor board meetings (MDTs). 4 An MDT is defined as a regularly scheduled meeting to review and set an integrated treatment plan for patients with cancer in the presence of experts such as surgeons, pathologists, radiologists, and medical and radiation oncologists. 5 MDTs have been introduced over time in cancer care facilities across many countries, and participation has been recommended to ensure prompt and adequate care by a variety of professionals. 6 MDTs have shown a positive effect on patient care and play a role in the planning of care. 7 For instance, 26% of patients with colorectal cancer had changes in their treatment plans after an MDT. 8 In another study, 29% of patients presenting with primary rectal cancer had changes in their treatments after their cases were discussed in an MDT. 9 Additionally, MDTs facilitated preoperative staging in 96% of patients with rectal cancer. 10 A systematic review showed that patients who were discussed at an MDT had a more accurate diagnosis and complete preoperative staging. The authors reported that the diagnostic reports changed in 4% to 35% of cases. 11 Patients presented to an MDT have better adherence to disease-specific guidelines and shorter time from diagnosis to treatment. 11,12 Patients who were discussed at an MDT had better survival than those who were not. 13 In another study, the 3-year survival rate in patients with advanced colorectal cancer was 66% in the MDT group vs 58% in the non-MDT group. 14 The 1-year survival rate in patients with non-small cell lung cancer discussed at an MDT was 33% compared with 18% in those who were not. 13 Previous literature suggests that further evaluation should be based on the quality and effectiveness of an MDT on treatment plans. 9,11,15 Despite the benefits of MDTs, there is limited evidence of the association between MDTs and mortality and in-hospital morbidity for mixed cases of gastrointestinal (GI) cancer. 11,15 In the absence of evidence, it is unclear whether optional MDTs should become a mandatory standard of care. Thus, the aim of this study was to measure the impact of an optional MDT on patients' mortality and in-hospital morbidity. We also aimed to examine disease-specific mortality among patients with colorectal, stomach, pancreatic, and hepatobiliary cancers.

| Formation of an MDT
An MDT was started in our center in January 2017. Our MDT includes the disciplines of medical oncology, gastroenterology, hepatology, surgery, diagnostic and interventional radiology, radiation oncology, and pathology. The MDT is optional, and the decision on whether a patient's case should be presented in the MDT is made by the primary treating physician.

| Variables and outcomes
Cancer recurrence and metastasis were defined as the prevalence of cancer occurrence before the last event of current cancer that was included in our final analysis as a baseline characteristic. For the sake of analyzing the outcomes, we combined colon and rectal cancer into the colorectal cancer (CRC) category and combined liver cancer, cholangiocarcinoma, and pancreatic cancer into the hepatobiliary and pancreatic (HBP) cancer category. We did not include patients with small bowel or esophageal cancer as we only had a total of three patients with these conditions and all of them were discussed in the MDT, with no patients in the control group (non-MDT). We considered patients who were not treated in our center as missing, and thus, we dropped them from our analysis. We categorized patients into two groups: the MDT group, if patients were discussed in an MDT, and the control group (non-MDT), if patients were not discussed in an MDT. The primary outcome was overall mortality, and the secondary outcome was disease-specific mortality for patients with stomach cancer, HBP cancer, and CRC.

| Statistical analysis
Demographic and clinical characteristics were compared between the two groups using a two-way t-test if the data were normally distributed or the Mann-Whitney test in case of non-normally distributed data. Fisher's exact test or the chi-squared test was used for cross tabulation of data. We defined the time to event as the time from diagnosis to death or end of the study. We constructed unadjusted Kaplan-Meier estimates to describe the failure function in comparing the non-MDT and MDT groups, and the log-rank and Wilcoxon tests were applied to test equality. Adjusted failure-function curves were constructed to describe the time to death between groups after adjusting for other factors. Univariable and multivariable Cox proportional hazards analyses were used to identify the independent factors for overall mortality. Multivariable Cox proportional hazards analysis was stratified by cancer type for subgroup analysis. Covariates in the models included male sex, BMI, nationality, ACCI score, and history of cancer recurrence, which were selected based on clinical relevance, prior literature, and the guidance of likelihood-ratio tests to achieve model parsimony. A P value of <.05 was considered statistically significant. Analyses were conducted using Stata 14 statistical software (StataCorp, College Station, TX, USA).

| Patients' demographic and clinical data
Over the study period, a total of 130 patients were included. Sixty-six patients were in the non-MDT group and 64 patients were in the MDT group. Patients' demographics and clinical data are summarized in Table 1. The non-MDT group had more patients with CRC (55, 83%) than the MDT group (41, 64%) (P = .02). In contrast, the non-MDT group had fewer patients with HBP cancer (4, 6%) than the MDT group (18, 28%) (P < .01). No significant differences were observed between the two groups with regard to age, body mass index (BMI), nationality, sex, history of cancer recurrence, ACCI score, or metastasis. Tables 2 and 3 shows no significant difference in treatment plans between groups, except that 21 (33%) patients in the non-MDT group were treated by a combination of chemotherapy, radiotherapy, and surgery compared with eight (13%) in the MDT group (P = .01). Six (9%) patients in the MDT group were treated with other therapies, for example, portal vein (PV) embolization, radio-frequency ablation (RFA), and imatinib, compared with one patient in the non-MDT group (P = .12).  (Table 4). There were no significant differences between the groups with regard to hospital length of stay, ICU length of stay, or time from diagnosis to surgery (Table 4).

| Overall mortality and specific mortality
Over the study period, the follow-up was 100% complete, with no patients censored. Four patients (6%) in the MDT group died compared with 13 (20%) in the non-MDT group (P = .04). No significant differences were found in overall mortality at 6 months, 1 year, and postoperative mortality (5%) than non-MDT patients (9%). 16 Furthermore, a study conducted among CRC patients showed that MDT was an independent factor for survival, as the 3-year survival post-MDT establishment was 66% compared with 58% in non-MDT patients. 14 Another study assessed the evolution of colorectal cancer treatment over the years in a single center, where the 3-year overall survival after the initiation of a multidisciplinary colorectal cancer center was dramatically increased, to 82% from 65%. 17 Several previous studies have looked into survival among patients with hepatocellular carcinoma before and after MDT establishment, 18,19 and all found improved survival post-MDT. One of these studies found a 28% reduction in mortality among MDT-discussed cases compared with non-MDT-discussed cases. 18 According to our results, the prevalence of CRC was lower and that of HBP cancer was higher in the MDT group than in the non-MDT group. This might be explained by the fact that CRC is ranked as the most common cancer in men and the third most common cancer in women in Saudi Arabia, as reported by the WHO; well-defined mortality. This might be due to the effectiveness of an MDT for common cancers such as CRC and rare cancers such as HBP cancer.
Unsurprisingly, we found that BMI had a role in the mortality of patients with cancer; mortality decreased by 12% with every 1 unit increase in BMI. This indicates that patients who are generally fit and healthy survive longer. Our finding is in line with that of a previous study that compared CRC patients with a normal BMI with those with changes in BMI; the study showed that for stage III disease, lower BMI was associated with an 87% increase in mortality. Furthermore, a protective effect was noted for stage IV disease, as there was a 42% decrease in mortality among those with a higher BMI. 20 Other evidence supporting our finding comes from Li et al, who found that for each 5% decrease in BMI, there was a 27% increase in overall mortality; they concluded that increased mortality was linked to lower BMI. 21 In contrast, Shaukat et al's study on CRC mortality found that a higher BMI was related to increased mortality. 22  increased significantly from 4.5 months to 9.5 months after its implementation. 24  There are other questions that our study was not designed to answer. First, we are aware that the mortality and prognosis in GI cancer differ according to the histopathological diagnosis and radiological staging, but we decided to look into the effectiveness of an MDT in general. For that, we did not collect data on histopathology, radiological staging, performance status, albumin, or lactate dehydrogenase.
Second, the change in treatment plans due to an MDT is well documented in the literature. However, we did not collect any data about the change in treatment plans, and our research team decided to focus on more meaningful clinical outcomes that reflect the patients' general health. Third, in our center, MDT discussions are optional, which places the decision of presenting the patients' cases in the hands of the primary treating physician. This, along with the nature of our study design, might introduce selection bias. Despite the fact that our center is considered one of the largest centers for treating cancer in the city of Jeddah, Saudi Arabia, this study is a single-center study with a small sample size, which might affect the generalizability of our findings and limit our conclusions to similar centers. To address these limitations in our study design, further research using a multicenter prospective study that includes a large sample size and homogeneous cohort with cancer-specific data (eg, data on radiological staging, performance status, albumin, and lactate dehydrogenase) is recommended.

| CONCLUSIONS
This study revealed that MDTs had a significant role in decreasing mortality, with no effect on morbidity. This may have been because of the well-structured and integrated management plan provided to MDT patients, making them and their treatment algorithm well known to the oncology services in the hospital. MDTs have a meaningful positive influence on patient care by improving survival and should be incorporated into the standard care of patients with cancer. We recommend including other disciplines in the MDT, for example, nurses and nutritionists, to improve patients' general health before treatment. A further multicenter prospective study that includes cancer-specific data (eg, data on radiological staging) is recommended.

ACKNOWLEDGMENT
We would like to thank Editage (www.editage.com) for English language editing.