Recurrence of Breast Cancer After Primary Treatment: A Matched Comparison Study of Disease-Free Survival in Women Who Do and Do Not Receive Adjunctive Naturopathic Oncology Care

Purpose:

To compare disease free survival experienced by women who received usual oncologic care compared to a cohort of women who received naturopathic oncology care in addition to usual care.

Methods:

Women with breast cancer who received naturopathic oncology (NO) care in Western Washington State (WA) (N = 176) were recruited to a prospective study of clinical health-related quality of life outcomes and then matched to women who received usual care (UC) only (N = 334).

Results:

Among 510 women with breast cancer stages 1 to 3, a total of 50 women (10%) experienced a disease-free survival (DFS) ending event within the observation period; 23 (6.8% of those in the UC cohort, and 27 (15.3% of those in the NO cohort (P < .05). Although, women in the 2 cohorts received similar surgical, chemotherapy, and radiation treatment, women with breast cancer who received naturopathic oncology adjunctive care were less likely to use anti-estrogen therapy, and experienced poorer DFS (logrank test, P < .05). However, differences in DFS could not be shown to be due to cohort differences in anti-estrogen therapy, baseline HRQOL, or naturopathic oncology therapies prescribed. The stage 3 women in the naturopathic oncology group had more advanced disease at diagnosis. They were more likely to have 5 or more metastatic lymph nodes at baseline (18.5%) compared to their usual care matched control group (13%). Women in the naturopathic oncology group also had higher grade tumors at diagnosis.

Conclusions:

Results show that recurrence of breast cancer was associated with more advanced malignant lymph node involvement; and that naturopathic oncology services provided in 2009-2015 did not improve disease-free survival in these high-risk breast cancer patients.

Building a predictive model for warfarin dosing via machine learning

Background

Warfarin titration via International Normalised Ratio (INR) monitoring can be a challenge for patients on long term anticoagulation due to its multiple drug interactions, long half life and patient's individual reponse, yet critically important due to its narrow therapeutic index. Machine learning may have a role in learning and replicating existing warfarin prescribing practices, to be potentially incorporated into an automated warfarin titration model.

Purpose

We aim to explore the feasibility of using machine learning to develop a model that can learn and predict actual warfarin titration practices.

Methods

A retrospective dataset of 4,247 patients with 48,895 data points of INR values were obtained from our institutional database. Patients who had less than 5 visits recorded, invalid or missing values were excluded. Variables studied included age, warfarin indication, warfarin dose, target INR range, actual INR values, time between titration and time in therapeutic range (TTR as defined by the Rosenthal formula).

The machine learning model was developed on an unbiased training data set (1,805 patients), further refined on a handpicked balanced validation set (400 patients), before being evaluated on two balanced test sets of 100 patients each. The test sets were handpicked based on the criteria of TTR (“in vs out of range”) and stability of INR results (“low vs high fluctuation”) (Table 1). Given the time series nature of the data, a Recurrent Neural Network (RNN) was chosen to learn warfarin prescription practices. Long-short term memory (LSTM) cells were further employed to address the problem of time gaps between warfarin titration visits which could result in vanishing gradients.

Results

A total of 2,163 patients with 42,622 data points were studied (mean age 65±11.7 years, 54.7% male). The mean TTR was 65.4%. The total warfarin dose per week as predicted by the RNN was compared with actual total warfarin dose per week prescribed for each patient in the test sets. The coefficient of determination for the RNN in the “in vs out of range” and “low vs high fluctuation” test sets were 0.941 and 0.962 respectively (Figure 1).

Conclusion

This proof of concept study demonstrated that a RNN based machine learning model was able to learn and predict warfarin dosage changes with reasonable accuracy. The findings merit further evaluation of the potential use of machine learning in an automated warfarin titration model.

Funding Acknowledgement

Type of funding source: None

Cancer beliefs in ethnic minority populations: a review and meta-synthesis of qualitative studies

People from ethnic minorities often experience poorer cancer outcomes, possibly due to later presentation to healthcare and later diagnosis. We aimed to identify common cancer beliefs in minority populations in developed countries, which can affect symptom appraisal and help seeking for symptomatic cancer. Our systematic review found 15 relevant qualitative studies, located in the United Kingdom (six), United States (five), Australia (two) and Canada (two) of African, African-American, Asian, Arabic, Hispanic and Latino minority groups. We conducted a meta-synthesis that found specific emotional reactions to cancer, knowledge and beliefs and interactions with healthcare services as contributing factors in help seeking for a cancer diagnosis. These findings may be useful to inform the development of interventions to facilitate cancer diagnosis in minority populations.

Vector-mediated expression of erythropoietin improves functional outcome after cervical spinal cord contusion injury

We evaluated the therapeutic effect of erythropoietin (EPO) delivered by direct injection of a nonreplicating herpes simplex virus (HSV)-based vector coding for EPO (vEPO) in a model of cervical hemicord contusion at C7. At 1 h after spinal cord injury (SCI), either vEPO or control vector carrying a reporter gene (vC) was injected into the cord above and below the lesion. Animals injected with vEPO showed a statistically significant improvement in the ipsilateral forelimb function, as measured by open-field evaluation of motor performance, forelimb reaching in the cylinder test and misplacement in grid walk. This correlated with preservation of gray matter in the area of the lesion. There was also mild but significant improvement of hindlimb motor function measured by Basso-Beattie-Bresnahan score and computerized gait analysis in vEPO compared with control vector-injected animals. Microtubule-associated protein tau, phosphorylated and nonphosphorylated neurofilament protein and the synaptic proteins synaptophysin and PSD-95 were all significantly increased in the spinal cord of vEPO-treated animals compared with control vector-injected animals. These data suggest that gene transfer of EPO after cervical SCI by minimizing the injury size and enhancing tissue sparing preserves large-caliber axons and promotes synaptogenesis.

MEPE's diverse effects on mineralization

Matrix extracellular phosphoglycoprotein (MEPE) is an inhibitor of mineralization in situ and in cell cultures where altered expression is associated with oncogenic osteomalacia and hypophosphatemic rickets. The purpose of this study was to determine whether the intact protein or the peptide(s) originating from this protein was responsible for the inhibition. The ability of the intact protein and the acidic, serine- and aspartate-rich MEPE-associated motif (ASARM) peptide to promote or inhibit de novo hydroxyapatite formation and growth of hydroxyapatite seed crystals, in both phosphorylated and dephosphorylated forms, was assessed at room temperature in a dynamic gel diffusion system at 3.5 and 5 days. The most effective nucleator concentration was also examined when associated with fibrillar type I collagen. The phosphorylated intact protein was an effective promoter of mineralization in the gelatin gel diffusion system, while the ASARM peptide was an effective inhibitor. When dephosphorylated both the intact protein and the ASARM peptide had no effect on mineralization. Associated with collagen fibrils, some of the effect of the intact protein was lost. This study demonstrates the importance of posttranslational modification for the site-specific activity of MEPE and its ASARM peptide.

Hedgehog signaling in germ cell migration

The primitive gonad of the Drosophila embryo is formed from two cell types, the somatic gonad precursor cells (SGPs) and the germ cells, which originate at distant sites. To reach the SGPs the germ cells must undergo a complex series of cell movements. While there is evidence that attractive and repulsive signals guide germ cell migration through the embryo, the molecular identity of these instructive molecules has remained elusive. Here, we present evidence suggesting that hedgehog (hh) may serve as such an attractive guidance cue. Misexpression of hh in the soma induces germ cells to migrate to inappropriate locations. Conversely, cell-autonomous components of the hh pathway appear to be required in the germline for proper germ cell migration.

The hypotensive action of 3-deaza-adenosine

The effects of 3-deaza-adenosine on the cardiovascular system was investigated in normotensive (WKY) and spontaneously hypertensive rate (SHR). This compound (50 mg/kg i.p. 2X) significantly lowers blood pressure in both groups of animals, an effect which lasts for up to 2 h. The hypotensive effect is not affected by ganglion blockade, and appears to be due in part to a direct effect on blood vessels. While the response is similar to that seem with adenosine, it is more long-lasting. The results of this study suggest a potential role for this compound or analogues of its as possible antihypertensive agents.