Applications of artificial intelligence (AI) in ovarian cancer, pancreatic cancer, and image biomarker discovery.

BACKGROUND: Artificial intelligence (AI), including machine learning (ML) and deep learning, has the potential to revolutionize biomedical research. Defined as the ability to "mimic" human intelligence by machines executing trained algorithms, AI methods are deployed for biomarker discovery. OBJECTIVE: We detail the advancements and challenges in the use of AI for biomarker discovery in ovarian and pancreatic cancer. We also provide an overview of associated regulatory and ethical considerations. METHODS: We conducted a literature review using PubMed and Google Scholar to survey the published findings on the use of AI in ovarian cancer, pancreatic cancer, and cancer biomarkers. RESULTS: Most AI models associated with ovarian and pancreatic cancer have yet to be applied in clinical settings, and imaging data in many studies are not publicly available. Low disease prevalence and asymptomatic disease limits data availability required for AI models. The FDA has yet to qualify imaging biomarkers as effective diagnostic tools for these cancers. CONCLUSIONS: Challenges associated with data availability, quality, bias, as well as AI transparency and explainability, will likely persist. Explainable and trustworthy AI efforts will need to continue so that the research community can better understand and construct effective models for biomarker discovery in rare cancers.

Establishing Trust in Pharmaceutical Data with an Independent Verification and Validation Methodology.

Our aim is to demonstrate a general-purpose data and knowledge validation approach that enables reproducible metrics for data and knowledge quality and safety. We researched widely accepted statistical process control methods from high-quality, high-safety industries and applied them to pharmacy prescription data being migrated between EHRs. Natural language medication instructions from prescriptions were independently categorized by two terminologists as a first step toward encoding those medication instructions using standardized terminology. Overall, the weighted average of medication instructions that were matched by reviewers was 43%, with strong agreement between reviewers for short instructions (K=0.82) and long instructions (K=0.85), and moderate agreement for medium instructions (K=0.61). Category definitions will be refined in future work to mitigate discrepancies. We recommend incorporating appropriate statistical tests, such as evaluating inter-rater and intra-rater reliability and bivariate comparison of reviewer agreement over an adequate statistical sample, when developing benchmarks for health data and knowledge quality and safety.

Impact of early dose intensity reduction of Palbociclib on clinical outcomes in patients with hormone-receptor-positive metastatic breast cancer.

BACKGROUND: Palbociclib is commonly added to an aromatase inhibitor (AI) as first-line therapy in ER + HER2- metastatic breast cancer (MBC). There are no data on the effect of the relative dose intensity (RDI) of palbociclib in first-line setting on clinical outcomes. The objective of this study is to explore the association of RDI and dose reduction of palbociclib in the first-line setting with PFS. METHODS: This is a retrospective study of ER + HER2- MBC patients who received palbociclib plus AI in first-line setting. Subjects ≥ 18 years old with MBC, who were started on palbociclib 125 mg daily, had completed ≥ 1 cycle of palbociclib, and did not progress within the first 12 weeks were eligible. Analyses were performed at 12- and 36-week landmarks (LM). RDI was defined as the total amount of palbociclib taken per the total amount planned. RDI-high-12 and RDI-low-12 cohorts were defined as patients receiving palbociclib with RDI ≥ 80% and RDI < 80% during the first 12 weeks, respectively. Reduction-12 and No-reduction-12 cohorts were defined as patients who had any dose reduction and patients who had no reduction during the first 12 weeks, respectively. RESULTS: 56 patients were eligible. Kaplan-Meier analysis from 12-week LM showed a median PFS of 17.1 months in RDI-high-12 versus 6.8 months in RDI-low-12 cohort (p = 0.0006). There was a 7.0-month improvement in median PFS in No-reduction-12 versus Reduction-12 cohort (p = 0.0638). Median PFS at 36-week LM was not reached in RDI-high-36 versus 8.6 months in RDI-low-36 cohort (p = 0.0703). CONCLUSIONS: RDI < 80% of palbociclib during the first 12 weeks, when used in combination with an AI in first-line setting in ER + HER2- MBC, is associated with significantly shorter PFS compared to RDI ≥ 80%. There is a trend towards shorter PFS among patients with RDI < 80% versus RDI ≥ 80% at 36 weeks. A larger study is needed to validate these findings.

Additional B-cell malignancies in patients with chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL).

Family and migration studies suggest a genetic risk of developing chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL). We hypothesized that CLL patients have an increased risk of additional clonally unrelated B-cell malignancies. To test this, we studied 467 CLL patients (2743 person-years (PYs)) at a single institution over 17 years. The incidence rate (IR) of any additional B-cell lymphoid malignancy was 10.9 per 1000 PYs (n = 30, 6.4%). Eighteen (4%) patients had a clonally unrelated B-cell malignancy (IR = 6.6 per 1000 PYs). Standardized incidence ratios (SIRs) were used to compare the incidence of additional clonally unrelated B-cell malignancies in CLL patients to the age- and sex-matched expected rates in the USA generated from the Surveillance, Epidemiology, and End Results (SEER) database. For the subset of 13 patients having data for comparison in the SEER database, the SIR was 5.41 (95% CI = 2.9, 9.3) which is supportive of our hypothesis.

Multiple B cell malignancies in patients with chronic lymphocytic leukemia: epidemiology, pathology, and clinical implications.

Chronic lymphocytic leukemia (CLL)/small lymphocytic lymphoma (SLL) is associated with increased risk for certain cancers, but relatively little is known about the risk for these patients to develop additional B cell malignancies. Here, we review the available epidemiological data on multiple B cell malignancies in CLL, discuss diagnostic methods and proper pathologic evaluation to distinguish CLL from other B cell malignancies, and address clinical challenges and unmet needs in caring for CLL patients with unrelated B cell malignancies and disease transformation. Considerations include CLL patients with unrelated monoclonal B cell lymphocytosis, biclonal CLL, secondary B cell non-Hodgkin lymphomas, and Richter syndrome - both clonally related transformation and de novo large B cell lymphoma. We address the challenges that remain in order to better understand the underlying risk factors and biology that may put CLL patients at increased risk of developing multiple B cell neoplasia.

Management of melanoma in patients with chronic lymphocytic leukemia.

Melanoma is significantly more common and is associated with a poorer prognosis in patients with an underlying B-cell malignancy. This study reports on the management of patients with chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL) and a subsequent diagnosis of melanoma. In the Wilmot Cancer Institute CLL cohort, which includes 470 patients followed for 2849 person-years, 18 patients (3.8%) developed 22 melanomas. Fourteen melanomas were invasive, a significantly higher rate as compared with the age and sex matched general population (standardized incidence ratio [SIR] 6.32 (95% CI 3.45; 10.60). Melanomas were most often detected (n = 15; 68.2%) through active surveillance in a dermatology clinic. Most melanomas (n = 17; 77.3%) were detected at a non-advanced stage (pathological stage grouping < III). The most common management was wide local excision without sentinel lymph node biopsy (n = 13, 59.1%). Management for the 4 (18.2%) patients with metastatic disease included the immune checkpoint inhibitor (ICI) pembrolizumab (n = 1), systemic chemotherapy with dacarbazine (n = 1), and palliative care (n = 2). The patient treated with ICI is in sustained remission of her melanoma after 23 cycles of therapy while her TP53 disrupted CLL continues to respond to ibrutinib therapy. We conclude that patients with CLL may benefit from active surveillance for melanoma leading to early excision of locally-manageable disease. In patients with metastatic melanoma, combined treatment with targeted kinase inhibitors and ICIs can be successful and tolerable. Larger prospective studies should be considered to further evaluate these approaches.

The association between early-life relative telomere length and childhood neurodevelopment.

PURPOSE: To examine the association between telomere length and neurodevelopment in children. METHODS: We examined the relationship between relative telomere length (rTL) and neurodevelopmental outcomes at 9 and 30 months, and 5 years of age in children enrolled in the Seychelles Child Development Study Nutrition Cohort 1 (NC1). Relative telomere length was measured in cord blood and in child blood at age five. Multivariable linear regression examined associations between neurodevelopmental outcomes and rTL adjusting for relevant covariates. RESULTS: Mean rTL was 1.18 at birth and 0.71 at age five. Increased cord blood rTL was associated with better scores on two neurodevelopmental tests, the psychomotor developmental index (ß = 4.01; 95% confidence interval (CI) = 0.17, 7.85) at age 30 months, and the Woodcock Johnson test of achievement letter-word score (ß = 2.88; CI = 1.21-4.56) at age five. The Woodcock Johnson test of achievement letter-word score remained statistically significant after two outliers were excluded (ß = 2.83; CI = 0.69, 4.97); the psychomotor developmental index did not (ß = 3.62; CI = -1.28, 8.52). None of the neurodevelopmental outcomes at age five were associated with five-year rTL. CONCLUSION: Although increased cord blood rTL was associated with better test scores for a few neurodevelopmental outcomes, this study found little consistent evidence of an association between rTL and neurodevelopment. Future studies with a larger sample size, longer follow-up, and other relevant biological markers (e.g. oxidative stress) are needed to clarify the role of rTL in neurodevelopment and its relevance as a potential surrogate measure for oxidative stress in the field of developmental neurotoxicity.

Interval robotic cytoreduction following neoadjuvant chemotherapy in advanced ovarian cancer.

The objective of this study is to review our experience with robotic interval cytoreduction following neoadjuvant chemotherapy for advanced ovarian cancer. We retrospectively reviewed patients with advanced ovarian cancer treated with neoadjuvant chemotherapy (NAC) and interval robotic cytoreduction (IRC) between 2011 and 2016 at the University of Rochester Medical Center. Demographic information, chemotherapy treatment, operative results, and follow-up were extracted from medical records. Twenty-nine patients underwent IRC after a mean of 3.9 cycles of NAC. The mean operative time was 165 min with a mean EBL of 107 cc. The mean length of stay was 2.0 days. One case (3.3%) was converted to an open procedure because of extensive tumor not amenable to robotic cytoreduction. Overall, 19 (66%) patients underwent an R0 cytoreduction, 8 (28%) an optimal (<1 cm) cytoreduction, and 2 (7%) a suboptimal cytoreduction. The median overall survival was 39.7 months and median progression-free survival was 21.2 months. Interval robotic cytoreduction following NAC is feasible and may be preferable to open interval cytoreductive surgery, in specific patients, to minimize morbidity and length of hospital stay.

Analysis of the risk of infection in patients with chronic lymphocytic leukemia in the era of novel therapies.

We studied the risk of infections in patients with chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL). Major infections were defined as requiring hospital admission or intravenous antimicrobial treatment. Incidence rate (IR) ratios (IRR) were used to compare infection rates. Of 263 CLL patients followed for 936.9 person-years, 60% required treatment for progressive CLL (66 received ibrutinib). Infections occurred in 71.9% patients (IR 92.4/100 person-years) with 31.9% having major infections (IR 20.3/100 person-years) and infections causing 37.5% of deaths. CLL treatment was associated with significantly higher risk of major (IRR 3.31, 95% CI 2.10, 5.21) and minor (IRR 1.78, 95% CI 1.43, 2.22) infections. Compared to their previous chemoimmunotherapy patients receiving salvage ibrutinib therapy (n = 47) had a significantly increased risk of a major infection (IRR 2.35 95% CI 1.27, 4.34). The risk of infection in CLL patients remains high even with use of less immunosuppressive therapies.

Late Relapses After High-dose Chemotherapy and Autologous Stem Cell Transplantation in Patients With Diffuse Large B-cell Lymphoma in the Rituximab Era.

BACKGROUND: The standard of care for diffuse large B-cell lymphoma (DLBCL) relapsing after front-line therapy is high-dose chemotherapy and autologous stem cell transplantation (ASCT). Evidence has suggested that early relapses (ie, within 1 year) after this approach portends exceptionally poor outcomes. However, data examining relapses > 1 year after ASCT for patients with refractory or relapsed DLBCL are limited, in particular, in the rituximab era. We sought to examine the effect of early (≤ 1 year) and late (> 1 year) relapse after ASCT in a single-institution cohort of patients with relapsed and refractory DLBCL treated with chemoimmunotherapy. MATERIALS AND METHODS: A retrospective analysis was performed on the data from 85 consecutive patients who had undergone ASCT for biopsy-confirmed relapsed or refractory DLBCL from 2001 to 2010 at the University of Rochester Medical Center. All patients had received rituximab as a part of treatment. Of the 85 patients, 35 developed relapse after ASCT. These 35 patients were divided into 2 groups according to the timing of the relapse (≤ 1 year and > 1 year after ASCT). RESULTS: The median follow-up period was 6.4 years. For all patients, the overall survival (OS) from post-ASCT relapse was 5.2 years. For the 27 patients developing relapse at ≤ 1 year after ASCT, the median OS was 0.6 year and progression-free survival was 0.4 year. For the 8 patients developing relapse at > 1 year after ASCT, the median OS was 5.9 years and progression-free survival was 2.9 years. CONCLUSION: Patients with relapsed or refractory DLBCL experiencing relapse > 1 year after ASCT had good outcomes. Despite the relative rarity in incidence, a significant risk of relapse of DLBCL after ASCT remains, suggesting the need for continued monitoring because of the possibility of later progression.