Utility of the "All of Us" Database for Radiation Oncology Research

PURPOSE/OBJECTIVE(S)

We provide an overview the All of Us Research Program, a National Institutes of Health funded research database, and report on the utility of this database for Radiation Oncology research.

MATERIALS/METHODS

The All of Us Research Program aims to create a large and diverse health database with participants from across the US. Patients consent to join the database and agree to share electronic health records, complete surveys, provide physical measurements, and donate at least one biospecimen. In this observational study, we used the public data browser feature of the All of Us Dataset to evaluate utility for future radiation oncology research. Within the database, we report on the number of cases of common cancers in the US, using ICD10 codes, and also the number of patients treated with common radiotherapy procedures, using CPT4 codes. We then qualitatively report on additional data of interest. Of note, public patient counts in All of Us is rounded to the nearest interval of 20. Additional tiers of data access exist, including individual level data, which were not used in this survey of public data.

RESULTS

The database includes 372,380 participants, of which 46,380 patients have a diagnosis of cancer. Within the identified top 10 cancers according to ACS, there were a total of 26,540 primary malignant cancers. The most common type of cancers were breast cancer (ICD10 - c50 [n = 6,960]), Prostate (ICD10 - c61 [n = 4,500]), Non-Hodgkin lymphoma (ICD10 - c85 [n = 2860]), Lung & bronchus (ICD10 - c34 [n = 2000]), Colon (ICD10 - c18 [n = 1940]), Melanoma of the skin (ICD10 - c43 [n = 1740]), Thyroid(ICD10 - c73 [n = 1720]), Kidney (ICD10 - c64 [n = 1300]), Urinary bladder (ICD10 - c67 [n = 1140]). The most common radiation therapy procedures were Computerized Tomography simulation (CPT - 77290 [n = 4,040]), Intensity Modulated Radiation Therapy (CPT - 77386 [n = 780]), 3D conformal radiation therapy (CPT - 77412 [n = 1,720]), Stereotactic Body Radiation Therapy (CPT - 77373 [n = 400]), Brachytherapy (CPT - 77770- 77772 [n = 240]), and Proton Therapy (CPT - 77520-77525 [n = 140]). Qualitatively, other data within All of Us include labs values, drug exposures, comorbidities, other procedures, physical measurements, and wearable biometrics (Fitbit) data. Genomic data typically includes germline testing using whole genome sequencing or genotyping array. Patient surveys include data on personal medical history, family health history, lifestyle, social determinants of health, COVID19 experience, and overall health. Potential outcome variables of interest to oncology research in the database include survival, cause of death, additional procedures or diagnoses after cancer treatment, and quality of life metrics derived from the survey instruments.

CONCLUSION

The All Of Us Research Database includes a large number of cancer cases, of which a substantial number received radiotherapy. Future work will focus on focused hypothesis-driven research questions.

Prediction of Progression After Cervix Cancer Radiotherapy Using a Machine-Learning Model on Pre-Treatment MRI

PURPOSE/OBJECTIVE(S)

MRI may be useful to identify women with cervical cancer at high risk of disease progression to test strategies of treatment intensification. The purpose of this study was to determine the value of a machine-learning model built on pre-treatment MRI for prediction of risk of progression after radiation therapy.

MATERIALS/METHODS

MagneticResonance Imaging (MRI) data for women with cervical cancer was collected from The Cancer Genome Atlas Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma Collection (TCGA-CESC) on the Cancer Imaging Archive (TCIA), which reported clinical, treatment, and imaging data from a single institution. 27 patients who had received radiation for cervical cancer were selected for input into a custom 3-D Residual Neural Network (ResNet) model with added custom layers specific to DICOM data in tensorflow python package. One T2 MRI per patient was used to predict recurrence free survival after radiation treatment, where patients were predicted to be "high risk" or "low risk" for disease recurrence as the output of the model. All slices of the T2 MRI were used. The model was validated using five-fold cross validation; 80% of the data was used to train each fold and 20% was used for testing. Final model statistical significance was confirmed through shuffle test at the p < 0.01 level. The clinical outcomes of patients and the model's "low-risk" and "high-risk" prediction were compared.

RESULTS

There were 27 patients in the study with mean age of 51 years (range 29-79). 20 patients had squamous cell carcinoma and 7 patients had adenocarcinoma. The stage breakdown consisted of 9 women IB, 2 IIA, 9 IIB, 2 IIIA, 2 IIIB, and 3 stage IV. 10 women were treated with radiation alone and 17 with chemo-radiation. 5 women received surgery in addition to radiation or chemoradiation. 21 patients received brachytherapy. Median follow-up of patients was 29 months (range 3-64). The model predicted 7 patients as "high risk" for recurrence; all 7 developed a recurrence during follow up. None of the 20 patients predicted to be "low risk" developed disease recurrence. Among all patients in the study, the two-year progression free survival (PFS) was 82.0%. Patients identified as "low risk" and "high risk" by model had two-year PFS of 100% and 43%, respectively. Among patients with recurrence, 3 developed local recurrence and 4 developed distant metastases. The ResNet model achieved cross-validated accuracy of 92% for prediction of progression-free survival (p<0.01).

CONCLUSION

A 3-D ResNet machine-learning model using pretreatment MRI image data can accurately predict clinical outcomes for cervical cancer following radiation therapy. Future work to confirm generalizability should focus on validation with a larger clinical dataset.

Theoretical and conceptual framework for a high school pathways to pharmacy program

OBJECTIVES

To determine whether participation in the University of Illinois at Chicago College of Pharmacy (UIC-COP) Pathways to Pharmacy, an early urban pipeline program, motivated underrepresented minority students to pursue a prepharmacy curriculum in college and choose pharmacy as a career.

METHODS

Over a 4-year period, underrepresented minority high school students participated in a comprehensive 6-week program that included 3 weeks of prepharmacy curriculum and intensive socialization and 3 weeks working as a pharmacy technician in a chain pharmacy. The High School Survey of Student Engagement (HSSSE) was administered 3 times to 120 program participants from 2005-2008, with 4 open-ended questions added to the pretest, 3 open-ended questions added to the test administered at the midpoint of the program, and 7 open-ended questions added to the posttest.

RESULTS

After completing the program, 88 (75%) of the 120 students enrolled in the college's prepharmacy curriculum and planned to pursue a career in pharmacy, 10 (8%) were not interested in pursuing a career in pharmacy, and 20 (17%) were undecided, compared to the pretest data which showed that 40 (33%) were interested in a career in pharmacy, and 80 (67%) were undecided (p < 0.0001).

CONCLUSIONS

Participation in a Pathways to Pharmacy program grounded in both a theoretical and conceptual socialization model framework increased the number of underrepresented minority students in the pipeline to pharmacy schools.

Biological barriers to therapy with antisense and siRNA oligonucleotides

Attaining the full therapeutic utility of antisense and siRNA oligonucleotides will require understanding of the biological barriers that stand between initial administration of these drugs and their final actions within cells. This review examines some of the key barriers that affect the biodistribution of oligonucleotides both in molecular form and when they are associated with nanocarriers. An understanding of the biological processes underlying these barriers will aid in the design of more effective delivery systems.

A phase I trial evaluating the safety and biological activity of iboctadekin (rhIL-18) in combination with rituximab in patients with CD20+ B-cell non-Hodgkin's lymphoma

8566

Background: Iboctadekin (rhIL-18) is an immunostimulatory cytokine that has demonstrated anti-tumor activity in several preclinical models. When administered as monotherapy in phase I clinical studies, rhIL-18 was safe, well tolerated and induced potent biological responses (e.g. Th1 cytokine production and expression of activation markers on NK, CD8+ and CD4+ cells). These data affirm the endogenous role of IL-18 as a co-stimulatory cytokine and suggest that its optimal use would be in a combination with other immune modulators such as rituximab. Methods: Patients with CD20+ B cell non-Hodgkin's lymphoma are being given rituximab (375 mg/m2) IV weekly for 4 consecutive weeks in combination with ascending doses of intravenous rhIL-18 (1 to 100 mcg/kg in 6 cohorts of 3 patients each) IV weekly for 12 weeks to identify a dose that is safe and tolerable and gives a maximum biological effect. Eligible patients must have disease which progressed after standard therapy or for which there is no effective standard treatment. Assessments include safety/tolerability, pharmacokinetics, pharmacodynamics (serum cytokines, peripheral blood phenotypic markers and tumor biomarkers), immunogenicity and anti-tumor activity. Results: To date, thirteen subjects have been enrolled in the first four cohorts (1, 3, 10 and 20 mcg /kg of rhIL-18). The combination is well tolerated with a safety profile similar to that observed with rituximab or rhIL-18 monotherapy. The pharmacodynamic response is as expected with a dose-dependent decrease in circulating activated (CD69+) NK cells within 4 hours after completing the rhIL-18 infusion which rebound to pre-dose levels within 2–4 days. Using the International Working Group response criteria for lymphoma, two subjects had complete responses at 10 and 20 mcg/kg, one subject had a partial response at 10 mcg/kg and three subjects had stable disease at 1, 3 and 3 mcg/kg. Conclusions: These data show that the combination of rhIL-18 and rituximab is safe, well tolerated and induces potent biological activity. This study will define the dose level to be used in a future phase II trial evaluating this combination in patients with relapsed or refractory follicular lymphoma.

[Table: see text]

Immunization of puppies in the presence of maternally derived antibodies against canine distemper virus

Vaccination of dams with modified-live canine distemper virus (CDV) vaccines will elicit high concentrations of colostral antibody, that although vital for protection of the pup during the first weeks of life, can interfere with active vaccination against the virus. In the present study, 12 pups, 7-9 weeks of age, with maternally derived immunity to CDV, were vaccinated with a canarypox-vectored CDV vaccine. These pups were protected against intravenous challenge with CDV. Three littermate pups that were unvaccinated all developed clinical signs of infection after challenge, and two of these control pups died.

Comparison of sotalol versus quinidine for maintenance of normal sinus rhythm in patients with chronic atrial fibrillation

Many clinicians choose sotalol for the prevention of recurrences of atrial fibrillation (AF) as an alternative to quinidine, which has been associated with an increase in long-term mortality. Using meta-analytic techniques, we compared the effects on maintenance of sinus rhythm and mortality of combined groups of patients with chronic AF treated with sotalol, quinidine, or a control drug. Rates of conversion at 6 months and mortality were combined for each group after performing sensitivity analysis to test for homogeneity. Bayesian estimates and corresponding 95% credibility intervals were constructed to compare the probabilities of achieving sinus rhythm and mortality among groups. A literature search revealed 4 sotalol studies, 6 quinidine studies, and 5 control studies that met inclusion criteria established a priori. The point estimates for maintaining normal sinus rhythm (at 6 months) and corresponding credibility intervals for the 3 groups were sotalol 50% (range 42% to 58%), quinidine 53% (range 48% to 59%), and control 32% (range 26% to 39%). When combining and comparing mortality effects, the following studies met the same inclusion criteria: 4 sotalol studies, 9 quinidine studies, and 7 control studies. The point estimates and corresponding credibility intervals for mortality in the 3 groups were sotalol 2.2% (range 0.6% to 4.8%), quinidine 3.0% (range 1.7% to 4.7%), and control 1.1% (range 0.3% to 2.4%). Sotalol and quinidine are comparable in their ability to maintain sinus rhythm at 6 months (about 50%) and both agents are superior to control. There is a trend for both agents to increase mortality with long-term therapy. These data do not support choosing sotalol over quinidine as a safer alternative for preventing recurrences of chronic AF.

The effect of propylthiouracil on subsequent radioactive iodine therapy in Graves' disease

OBJECTIVE

Antithyroidal drugs (ATD) are used in the management of Graves' disease either as primary therapy for several months while awaiting remission of the disease or as pretreatment for several weeks prior to definitive radioactive iodine therapy (RAI). We have reported previously that pretreatment with propylthiouracil (PTU) before definitive RAI therapy is associated with a higher RAI treatment failure rate than RAI therapy alone. The objectives of the current study were 2-fold. First, to verify the results of our prior study regarding the effect of PTU used as pretreatment before RAI in a cohort of patients from a different institution and, secondly, to better define the relationship between the number of days off PTU before RAI therapy and therapeutic efficacy of RAI dosing.

DESIGN

A retrospective review of Graves' disease patients treated from 1980 to 1994.

PATIENTS

Study patients had to meet the following inclusion criteria: radionuclide studies and thyroid hormone values consistent with Graves' disease, at least 1 year of follow-up data available and discontinuation of the ATD at least 4 days before RAI administration. Exclusion criteria included therapy with any ATD other than PTU or ATD therapy during or following RAI dosing.

MEASUREMENTS

Effectiveness of RAI therapy, days on PTU, days off PTU and calculated RAI dose to the thyroid were recorded for each subject. We compared the efficacy of RAI therapy in patients treated with PTU (used either as pretreatment in preparation for RAI therapy or as primary long-term therapy) before RAI administrations to those treated with RAI alone with special attention to the number of days on and off PTU before RAI dosing. Patients were considered RAI treatment failures if a second dose of RAI was required to achieve a euthyroid or hypothyroid state.

RESULTS

One hundred and sixteen patients met our study criteria. Forty patients received PTU therapy for a mean of 221 +/- 59 days. The PTU was discontinued for a mean of 60 +/- 25 days before RAI dosing. Persistent hyperthyroidism was seen in 9% (7/76) of patients treated with RAI alone. The failure rate of a single dose of radioactive iodine was significantly increased when PTU was discontinued between 4 and 7 days before the administration of RAI (29% vs 9% for RAI alone, P = 0.039). PTU discontinued for at least 1 week before RAI dosing was associated with a nearly 2-fold increase in failure rate, but this difference did not achieve significance (17% vs 9% for RAI alone, P = 0.24). Examining only those patients receiving PTU, patients who had successful single dose RAI therapy tended to receive a higher dose of RAI than patients failing RAI therapy (480 +/- 30 vs 410 +/- 40 MBq administered dose, P = 0.18; and 8.0 +/- 0.9 vs 5.5 +/- 1.1 MBq/g thyroid tissue calculated dose, P = 0.21). Furthermore, total serum thyroxine at diagnosis was significantly higher in patients failing RAI therapy after PTU administration than in patients successfully treated with RAI after receiving PTU (316 +/- 40 vs 225 +/- 13 nmol/L, P = 0.03).

CONCLUSIONS

Propylthiouracil discontinued 4-7 days before radioiodine dosing is associated with a significant increase in the failure rate of a single dose of radioiodine. Discontinuation of the propylthiouracil for at least a week before radioiodine administration is associated with a higher, although not statistically significant, radioiodine failure rate. In patients that require treatment with propylthiouracil before radioiodine therapy, a higher total serum thyroxine level at diagnosis is associated with an increased rate of radioiodine failure. Consideration should be given to increasing empirically the dose of radioiodine administered to Graves' disease patients that have received propylthiouracil within a week of radioiodine administration in an effort to decrease the radioiodine failure rate to an acceptable level.

Identification of exons in a novel embryonal carcinoma locus using the GRAIL program

Embryonal carcinoma (EC) cells have proven to be of particular value in studies of both oncogenesis and mammalian development, as well as in evaluating the relationship between these two phenomena. Infection of the EC cell line, NR1-0, with a defective retrovirus containing a neomycin resistance cassette (Neo(r)), produced a mutant cell line: NR1-6. Genetic analysis of this variant cell line indicates that there is only a single insertion site. Interestingly, however, the NR1-6 cell line is unique in its morphology, tumorigenicity, and differentiative potential (1). We have sequenced over 18 kb from the regions flanking the retroviral insertion which we then analyzed using the computer programs GCG and BLAST. Although homology was found to 4 B1 repeat elements (approximately 150 bp long) and a novel CA/GT dinucleotide repeat, no homology was found to any known genes (2). Furthermore, attempts to identify potential exons or transcripts using various molecular techniques and the above mentioned computer programs were all negative. Most recently we employed the GRAIL (Gene Recognition and Analysis Internet Link) computer program which was specifically designed to identify potential exons (3). Analysis with this program identified 5 exon candidates: two characterized as excellent (>90% probability) and three as marginal (>60% probability). Using reverse transcriptase (RT)-PCR we have demonstrated that the two 'excellent' and one of the 'marginal' exon candidates identified by GRAIL are expressed as mRNA in the mutant cells. Sequencing of these PCR products indicates that the mRNA is identical to the genomic DNA sequence. Thus, we have found that GRAIL provides an efficient, reliable means of identifying real exons within long regions of novel genomic DNA.

Novel mouse repetitive element structures in an embryonal carcinoma mutant cell line

A mutant embryonal carcinoma (EC) cell line, NR1-6, has been isolated subsequent to infection with a retroviral insert containing a neomycin resistance cassette (Neo(r)) (1). This variant cell line has only a single insertion site, but is unique in its morphology, tumorigenicity, and differentiative potential (2). In order to understand the exact nature of the mutational event, we have sequenced the regions immediately flanking the insertion site (encompassing over 18 kb). Searches for homology in GenBank using the GCG and Blast programs have revealed homology to only a few small conserved regions (4 B1 repetitive elements approximately 150 bp long and a CA/GT dinucleotide repeat). Interestingly, B1 repetitive elements have been reported to play a role in germ cell differentiation. One of these elements is immediately 5' upstream of the insertion site and it appears as if the retroviral insertion may be responsible for activating or repressing the expression of this element. The CA/GT repeat has not yet been mapped within the mouse genome. Thus, we have identified a novel mouse locus which apparently regulates a number of cellular phenotypes.

Restricted environmental stimulation technique improves human performance: rifle marksmanship

This study controlled for relaxation and guided imagery confounds noted in much previous research on enhancement of human performance using the restricted environmental stimulation technique (REST). Dry flotation REST was used where subjects lay ("floated") on a salt-water-filled bladder in a sound-attenuated, light-free chamber. 9 men and 3 women in a rifle marksmanship training course, exposed to dry-flotation REST, showed significantly higher rifle marksmanship scores than the university students who as matched controls were exposed to relaxation (9 men and 3 women). Further, only the former showed a significant pre- to posttest improvement in scores, which suggests REST's positive effects on marksmanship go beyond the induction of relaxation by hypnosis. The results support hypotheses summarized in 1982 by Barabasz regarding potentiation by REST of internally generated imagery and subsequent improvement observed in a nonREST posttest environment.

Exercise induced fatal sinusoidal ventricular tachycardia secondary to moricizine

Moricizine has been touted as having a low incidence of proarrhythmic effects. We present a case of proarrhythmia from moricizine, which presented as exercise induced ventricular tachycardia, and review the literature suggesting that this antiarrhythmic drug shares the proarrhythmic profile of other agents with predominant type Ic action. We conclude that moricizine has certain clinical and electrophysiological features that resemble type Ic antiarrhythmic agents. Precautions similar to those used when prescribing other drugs of this type should be followed when prescribing moricizine, including predischarge exercise testing.

Clonal analysis of myelodysplastic syndrome: monosomy 7 is expressed in the myeloid lineage, but not in the lymphoid lineage as detected by fluorescent in situ hybridization

Conflicting results have been published on whether or not myelodysplastic syndromes (MDS) affect all cell lineages. Involvement of myeloid and erythroid cell lineages has been regularly observed, but it remains controversial whether the different lymphoid cell lineages are involved. In this study of eight patients with MDS associated with monosomy 7, fluorescent in situ hybridization (FISH) was used to enumerate the chromosomes 7 in interphase cells. With the probe D7Z1, the rate of false-positive detection of monosomy 7 was 3% +/- 2% in normal cells. T- and B-cell lines were established from eight patients with MDS and monosomy 7. As determined by FISH in interphase cells, 1.9% (0% to 3%) of the cells in the B-cell lines showed one fluorescent spot and 1.1% (0% to 2.9%) of the cells in the T-cell lines. These values do not differ from normal values. However, the possibility that normal cells were selected when the T- and B-cell lines were established could not be excluded. Therefore, peripheral blood cells were obtained, separated according to surface markers specific for lymphoid and myeloid cell lineage with a cell sorter, and analyzed for the expression of monosomy 7 by FISH. Antibodies recognizing T cells (CD3), B cells (CD20), natural killer (NK) cells (CD57), monocytes and granulocytes (low and high expression of CD11b antigen), and myeloid progenitors (CD33) were used to separate cells. The expression of monosomy 7 in the T cells, NK cells, and B cells did not differ from control values. These results in the lymphoid subpopulations are in stark contrast with the observations in the myeloid populations; the percentage of cells with monosomy 7 ranged from 9% to 78% (controls: 6% +/- 2%) in cells with low CD11b expression, 20% to 89% in cells with a high expression of the CD11b antigen (controls: 7% +/- 3%), and 23% to 91% in the CD33 positive cells (controls: 5% +/- 3%). The results of this study suggest that monosomy 7 does not usually affect lymphoid subpopulations but is restricted to committed progenitor cells with the capacity to differentiate into mature myeloid cells.