AI-Based Liver Allocation More Equitable Than MELD Score
Liver transplant candidates may get a fairer shake from an artificial intelligence-based organ allocation system than that offered by the current Model for End-Stage Liver Disease (MELD) scoring system, investigators contend.
A retrospective analysis of data from candidates on the waitlist for a deceased-donor liver indicated use of a machine-learning model trained to predict the probability of a candidate's death or removal from the list within 3 months would have resulted in an average annual reduction in deaths of 418 patients compared with the MELD score.
In simulations, the artificial intelligence model, dubbed Optimized Prediction of Mortality (OPOM), was associated with improved survival across all candidate demographics, geographic regions, and diagnoses. OPOM was substantially more accurate than the MELD score at predicting risk across all disease severity groups, according to Dimitris Bertsimas, PhD, from the Massachusetts Institute of Technology in Cambridge, and colleagues.
The researchers published their findings online November 9 in the American Journal of Transplantation.
"The application of an OPOM-based allocation system would more accurately adhere to the 'sickest-first' principle. Indeed, the decrease in waitlist mortality/removal achieved through utilization of OPOM would not only represent the potential for more equitable allocation, but also would represent an important facet towards alleviating the discrepancy between supply and demand," they write.
The MELD score has been used since 2002 to rank liver transplant candidates by disease severity, but the system's method of "exception points," which are intended to account for patients at imminent risk for death or disease progression, has resulted in what the authors called "inequitable and undesirable outcomes."
Specifically, the exception point policy gives too much weight to candidates with hepatocellular carcinoma at the expense of candidates without exception points, the researchers maintain.
The investigators sought to determine whether a machine-learning approach using a technique known as Optimal Classification Tree modeling could be better than the MELD score at answering the following question: "What is the probability that a patient will either die or become unsuitable for liver transplantation within 3 months, given his or her individual characteristics?"
They applied the OPOM to data from the Organ Procurement and Transplantation Network Standard Transplant Analysis and Research dataset, including information on patients on the waitlist from January 1, 2002 through September 5, 2016.
The researchers first trained the system to predict the probability of a patient dying or becoming unsuitable for transplant within 3 months as a dependent variable, given observations of certain patient characteristics as independent variables. The independent variables included demographic and clinical characteristics.
After applying the trained model to the data, the researchers determined that liver allocation according to OPOM scores would have resulted in 417.96 (17.6%) fewer deaths annually among patients on the waitlist compared with the Match MELD (ie, MELD with exceptions) score. Additional analysis showed that OPOM would reduce deaths compared with MELD across all United Network for Organ Sharing regions.
"Notably, a higher number of female candidates received transplants when OPOM allocation was utilized," the researchers write.
Compared with the Match MELD score, the OPOM score would have decreased deaths among patients on the waitlist, patients removed from the list, and post-transplant by 23.3%, 21.5%, and 1.8%, respectively.
Although OPOM allocated more livers to patients without hepatocellular carcinoma than MELD, OPOM decreased both waitlist deaths and list removals for patients with and without hepatocellular carcinoma.
The OPOM model "considerably outperformed" MELD for all patient exception statuses at predicting the 3-month probability of death or becoming unsuitable for transplant, as evidenced by a higher area-under-the curve of receiver operating characteristics.
"OPOM more accurately and objectively prioritizes candidates for liver transplantation based on disease severity, allowing for more equitable allocation of livers with a resultant significant number of additional lives saved every year. These data demonstrate the potential of machine learning technology to help guide clinical practice, and potentially guide national policy," the researchers write.
The study had no specified funding. The researchers have reported no relevant financial relationships.
Am J Transpl. Published online November 9, 2018. Abstract
Should We Cure Hepatitis C Virus in Patients With Hepatocellular Carcinoma While Treating Cancer?
Abstract and Introduction
Direct acting antivirals stabilize or improve liver function in the majority of patients with hepatitis C virus cirrhosis. Hepatic decompensation is the main driver of death of patients with early, successfully treated hepatocellular carcinoma superimposed to cirrhosis. Treatment with direct acting antivirals could improve the prognosis of these subjects, independently from the subsequent course of hepatocellular carcinoma, if the efficacy in obtaining viral clearance is as high as in patients without a history of hepatocellular carcinoma, and if the risk of hepatocellular carcinoma recurrence is unaffected. When dealing with hepatocellular carcinoma patients, direct acting antivirals can be indicated in two different settings: (a) subjects in which hepatocellular carcinoma has been already successfully treated ("cured" hepatocellular carcinoma), or (b) subjects whose hepatocellular carcinoma is still untreated or untreatable ("active" hepatocellular carcinoma). Although there are abundant data on "cured" hepatocellular carcinoma, evidence supporting treatment decisions in patients with "active" hepatocellular carcinoma is at best scarce and controversial, since these patients as well as patients with hepatocellular carcinoma listed for liver transplantation are usually excluded from treatment.
Hepatocellular carcinoma (HCC) is a major health care problem with an increasing incidence worldwide and a poor prognosis, and is the leading cause of mortality in patients with cirrhosis.[1,2] Cirrhosis is the strongest risk factor for HCC, with hepatitis C virus (HCV) being a major risk factor in the Western world and Japan.[3,4] The prognosis of patients with cirrhosis because of HCV is driven by the progression towards hepatic decompensation and HCC, the latter being the leading cause of mortality in patients with compensated cirrhosis.[5,6] On the other hand, the impairment of liver function has a significant impact on prognosis of patients with HCC, irrespective of the tumour stage. Moreover, hepatic function defines the capacity to indicate treatments with potential deleterious effects on the liver.[3,7]
Treatment of HCV was revolutionized by the advent of the new direct acting antivirals (DAAs) with high rates (>90%) of sustained virological response (SVR) even in advanced cirrhosis, modest contraindications and a low rate of adverse events. Randomized trials assessing DAAs for registration purposes have, however, excluded intentionally patients with HCC, hence the available data regarding these patients come from postmarketing surveillance, and mostly concern occurrence (ie, "de novo" HCC) or recurrence (relapse of HCC) rates during and after DAA treatment of patients with HCV cirrhosis. The issue of antiviral efficacy and of effects on the course of prognosis of these patients was not tackled by these studies.
Hepatic decompensation is the main driver of death of patients with early, successfully treated HCC superimposed to cirrhosis. Treatment with DAAs could improve the prognosis of these subjects, independently from the subsequent course of HCC, if the efficacy in obtaining viral clearance is as high as in patients without a history of HCC, and if the risk of HCC recurrence is unaffected.
In 2016, an alarm signal was released about a potentially increased risk of HCC (both "de novo HCC" in cirrhotic patients without history of HCC, and "recurrent" in patients successfully treated for HCC), after DAA therapy.[9,10]
Even if recent data from retrospective and prospective large cohort studies definitively clarified that SVR is associated with a reduction in the risk of incident HCC in the DAA era and that cirrhosis is the main driver of HCC risk after SVR,[11,12] doubts remain regarding the risk of HCC recurrences in successfully treated HCC (mainly because of methodological limitations of the studies) and very few data there are on patients with untreated HCC.
This review will focus on the current state of knowledge in HCV treatment of patients with successfully treated and untreated HCC and on the impact of viral clearance achieved with DAAs in the setting of patients with HCC and decompensated cirrhosis and of patients with HCC listed for liver transplantation (LT).
What Role Will Avelumab Play in Lung Cancer?
The anti–programmed death ligand 1 (PD-L1) agent avelumab did not improve overall survival compared with docetaxel in a phase III trial of patients with PD-L1–positive non–small-cell lung cancer (NSCLC) who had already received platinum-based therapy.
“Antibodies targeting the immune checkpoint molecules PD-1 or PD-L1 improve overall survival compared with standard-of-care chemotherapy in patients with metastatic NSCLC,” wrote study authors led by Fabrice Barlesi, MD, of Aix Marseille University in France. More recent studies have expanded some of those agents’ approvals into earlier NSCLC settings.
The new study compared avelumab, an anti–PD-L1 agent, with docetaxel in patients with stage IIIB or IV NSCLC; all patients had progressed after platinum-based therapy. The results were published in Lancet Oncology.
The JAVELIN Lung 200 study included a total of 792 patients treated at 173 hospitals in 31 countries. They were randomized to receive either avelumab (396 patients, 264 PD-L1–positive tumors) or docetaxel (396 patients, 265 PD-L1–positive tumors); the primary analysis was only of the PD-L1–positive patients. Just over two-thirds of the study population was male, and a similar proportion was white. Approximately 90% of the study participants had received one prior line of therapy.
The median overall survival did not significantly differ between the groups, at 11.4 months with avelumab and 10.3 months with docetaxel, for a hazard ratio (HR) of 0.90 (96% CI, 0.72–1.12; P = .16).
PD-L1 positivity was defined as expression of PD-L1 in at least 1% of tumor cells; when this was adjusted to higher cutoffs, avelumab did have greater efficacy. In patients with at least 50% expression (168 patients in the avelumab group, 147 in the docetaxel group), the median OS was 13.6 months with avelumab and 9.2 months with docetaxel, for an HR of 0.67 (95% CI, 0.51–0.89; P = .0052). At at least 80% expression (120 and 106 patients, respectively), the median OS was 17.1 months and 9.3 months, respectively, for an HR of 0.59 (95% CI, 0.42–0.83).
Grade 3–5 adverse events occurred in 10% of avelumab patients and in 49% of docetaxel patients. With avelumab, the most common such adverse events included infusion-related reactions and increased lipase, while with docetaxel the most common included neutropenia, febrile neutropenia, and decreased neutrophil counts. Serious adverse events were more common with docetaxel, and there were 4 treatment-related deaths with avelumab compared with 14 with docetaxel.
“Overall, although this trial did not meet its primary endpoint, the clinical activity and safety noted in this study support further studies of avelumab in patients with NSCLC,” the authors wrote.
Sameep Sehgal, MD, of Temple University’s Lewis Katz School of Medicine in Philadelphia, who was not involved with the research, said the results “do give us pause,” and that clinicians should remember that not all immune therapeutic agents are the same. “The results may dampen the enthusiasm about using avelumab in advanced lung cancer,” he told Cancer Network, though he pointed out that other studies have shown improved survival with other immunotherapeutic agents. “Improved survival in patients with higher PD-1 expression is encouraging and avelumab may have a role in this subgroup of patients in the future.”
SAN ANTONIO -- Aggressive, targeted radiation therapy delayed cancer progression and showed a strong signal toward improving overall survival (OS) in patients with oligometastatic disease, the SABR-COMET study found.
At a median 27 months follow-up, progression-free survival (PFS) was 12 months in patients treated with standard palliative care plus stereotactic body radiation therapy (SBRT) compared with 6 months in those who received palliative care alone (P=0.001), reported David Palma, MD, PhD, of the London Health Sciences Centre in London, Ontario, Canada.
And at 5 years, 46% of patients in the SBRT arm were alive compared with 24% in the control arm. Median OS was numerically better in SBRT-treated patients, at 41 months (95% CI 26 months to not reached) compared with 28 months (95% CI 19 to 33 months) in the control group (P=0.09).
Most of the patients had oligometastatic (≤3 lesions) cancers of the prostate, breast, colon, and lung.
Results of the trial will be presented next week in a plenary session at the American Society for Radiation Oncology (ASTRO) meeting here.
"It's been a big source of controversy as to how to best manage these patients," Palma told MedPage Today. "For many years people have suggested that maybe we should use aggressive treatments like surgery or stereotactic radiation where others said that kind of approach would be futile."
Palma said that he hopes this study, which will continue to follow patients for up to 10 years, will help to reduce that level of controversy. The trial was designed to detect a signal of benefit (P<0.20) but the PFS benefit was found to be definitive.
"The results are very tantalizing in that we're getting 13 months of difference in the survival, and the percentage of patients alive at 5 years is roughly double," said Palma.
"At our center, it's changed how we are practicing," he said, but noted that without the significant improvement in OS, decisions will ultimately come down to the individual oncologist and his or her discussions with patients.
"Many people feel that a progression-free survival benefit is enough to change practice, but others will want more information," he added.
"This study has very exciting implications for patients with metastatic disease, in particular for those with only a couple sites of metastases," James B. Yu, of Yale Cancer Center and Smilow Cancer Hospital in New Haven, Connecticut, told MedPage Today in an email.
"I can envision a future where ever more sophisticated molecularly-targeted therapies and immunotherapies in combination with precisely targeted and non-invasive radiosurgery lead to the cure of previously incurable metastatic disease," said Yu, who was not involved in the study. "There will always be a role for locally directed therapy -- and as radiosurgery gets more precise (with improvements in patient tracking and tumor visualization) and easier to plan and deliver (with artificial intelligence-driven improvements in radiation treatment planning), treatment of multiple lesions will become commonplace."
The open-label SABR-COMET trial enrolled 99 patients from 2012 to 2016 with recurrent cancer and good performance status (ECOG 0-1) from Australia, Canada, Scotland, and the Netherlands. Among the 99 patients, there were 16 with prostate cancer, and 18 with breast, lung, and colorectal cancers each. Patients with up to five metastases were included in the trial, but 93% had between one and three. Some patients who developed additional lesions during the trial were successfully treated with additional ablation.
Palma said that patients with different cancer types were included because SBRT works fairly well regardless of histology and other trials have failed to fully accrue in this setting.
The two groups of patients -- median age 68 and 59% men -- had similar baseline characteristics and were randomized 2:1 to SABR plus standard palliative care or palliative care alone. Adverse events (AEs) grade ≥2 were more common with SBRT (30% vs 9%). Most comment events were fatigue, dyspnea, muscle, joint, bone, or other types of pain. Three deaths occurred in the investigational arm due to AEs.
No differences were seen with regard to quality of life. At 6 months post-treatment, overall scores on the Functional Assessment of Cancer Therapy General (FACT-G) questionnaire were high in both treatment arms: 82.5 with SBRT versus 82.6 (P=0.992). And no significant differences were seen on the questionnaire's functional, emotional, physical, and social subscales.
A follow-up phase III study, SABR-COMET-3, will test the approach in patients with up to three metastatic lesions, and the phase III SABR-COMET-10 will test this approach with lower SBRT doses in patients with up to 10 lesions.
The study was funded by the Ontario Institute for Cancer Research and a London Regional Cancer Program Catalyst grant.