Milestones in the natural course of the onset of e-cigarette dependence among adolescents and young adults: Retrospective study

INTRODUCTION

There is considerable controversy about the development of dependence among e-cigarette users. This study describes the average amount of time using e-cigarettes before dependence milestones emerge and the differences in developing dependence between e-cigarette users who smoke cigarettes compared to those who do not.

METHODS

Adolescents and young adults aged 16-25 living in Canada were recruited into an online survey in 2021. Current (past-month) e-cigarette users completed 15 items assessing dependence from the Penn-State Electronic Cigarette Dependence Index and the E-cigarette Dependence Scale for a total of 12 behavioural (e.g., difficulty refraining from vaping) and 3 frequency (e.g., using e-cigarette daily, weekly, or monthly) indicators of dependence milestones. Number of years after e-cigarette onset at which the cumulative probability of attaining each milestone was 25 % was computed.

RESULTS

Among 1205 participants, most (80.6 %) were female, 73.7 % were Caucasian, and 49.7 % resided in Ontario. Ten of the 12 e-cigarette use milestones were attained by 25 % of respondents 2 years after starting vaping except for daily cigarette use (2.5 years after onset) and waking at night to vape (5.6 years after onset). Within the entire study population, frequency milestones (weekly, monthly, daily e-cigarette use) were attained faster by ever-smokers (hazard ratio compared to attainment by never-smokers: 1.12, 1.21, and 1.28 respectively), whereas for at least monthly users, behavioural milestones were attained faster by never-smokers.

DISCUSSION

Many current e-cigarette users developed symptoms of e-cigarette dependence between two and five years since onset. Never smokers may be at higher risk of becoming e-cigarette dependent since they attained e-cigarette dependence milestones faster than smokers.

Are intravitreal injections essential during the COVID-19 pandemic? Global preferred practice patterns and practical recommendations

Tertiary outpatient ophthalmology clinics are high-risk environments for COVID-19 transmission, especially retina clinics, where regular follow-up is needed for elderly patients with multiple comorbidities. Intravitreal injection therapy (IVT) for chronic macular diseases, is one of the most common procedures performed, associated with a significant burden of care because of the vigorous treatment regimen associated with multiple investigations. While minimizing the risk of COVID-19 infection transmission is a priority, this must be balanced against the continued provision of sight-saving ophthalmic care to patients at risk of permanent vision loss. This review aims to give evidence-based guidelines on managing IVT during the COVID-19 pandemic in common macular diseases such as age-related macular degeneration, diabetic macula edema and retinal vascular disease and to report on how the COVID-19 pandemic has affected IVT practices worldwide.

To illustrate some real-world examples, 18 participants in the International Retina Collaborative, from 15 countries and across four continents, were surveyed regarding pre- and during- COVID-19 pandemic IVT practices in tertiary ophthalmic centers. The majority of centers reported a reduction in the number of appointments to reduce the risk of the spread of COVID-19 with varying changes to their IVT regimen to treat various macula diseases. Due to the constantly evolving nature of the COVID-19 pandemic, and the uncertainty about the normal resumption of health services, we suggest that new solutions for eye healthcare provision, like telemedicine, may be adopted in the future when we consider new long-term adaptations required to cope with the COVID-19 pandemic.

Two-color Kerr microscopy of two-dimensional materials with sub-picosecond time resolution

We present a two-color Kerr microscopy system based on two electronically synchronized erbium-fiber laser oscillators with independently tunable emission energies spanning most of the visible spectrum. Combining a spatial resolution below 2 μm and sub-ps time resolution with high sensitivity and cryogenic sample temperatures, it is ideally suited for studying spin and valley dynamics in a wide range of two-dimensional materials. We illustrate its capabilities by studying a monolayer of the common semiconducting transition metal disulfide MoS₂.

Adaptive control of laser-wakefield accelerators driven by mid-IR laser pulses

There has been growing interest both in studying high intensity ultrafast laser plasma interactions with adaptive control systems as well as using long wavelength driver beams. We demonstrate the coherent control of the dynamics of laser-wakefield acceleration driven by ultrashort (∼ 100 fs) mid-infrared (∼ 3.9 μm) laser pulses. The critical density at this wavelength is 7.3 × 10¹⁹ cm^-3, which is achievable with an ordinary gas target system. Interactions between mid-infrared laser pulses and such near-critical-density plasma may be beneficial due to much higher absorption of laser energy. In addition, the normalized vector potential of the laser field a₀ increases with longer laser wavelength, lowering the required peak laser intensity to drive non-linear laser-wakefield acceleration. Here, MeV level, collimated electron beams with non-thermal, peaked energy spectra are generated. Optimization of electron beam qualities are realized through adaptive control of the laser wavefront. A genetic algorithm controlling a deformable mirror improves the electron total charge, energy spectra, beam pointing and stability at various plasma density profiles. Particle-in-cell simulations reveal that the optimal wavefront causes an earlier injection on the density up-ramp and thus higher energy gain as well as less filamentation during the interaction, which leads to the improvement in electron beam collimation and energy spectra.

Laser wakefield acceleration with mid-IR laser pulses

We report on, to the best of our knowledge, the first results of laser plasma wakefield acceleration driven by ultrashort mid-infrared (IR) laser pulses (λ=3.9  μm, 100 fs, 0.25 TW), which enable near- and above-critical density interactions with moderate-density gas jets. Relativistic electron acceleration up to ∼12  MeV occurs when the jet width exceeds the threshold scale length for relativistic self-focusing. We present scaling trends in the accelerated beam profiles, charge, and spectra, which are supported by particle-in-cell simulations and time-resolved images of the interaction. For similarly scaled conditions, we observe significant increases in the accelerated charge, compared to previous experiments with near-infrared (λ=800  nm) pulses.

Novel morphological macular findings in juvenile CLN3 disease

AIMS

Juvenile CLN3 disease, one of the most common forms of a group of lysosomal storage diseases called neuronal ceroid lipofuscinoses (NCLs), is a progressive neurodegenerative disorder with initial visual deterioration. The objective of this study was to analyse the retinal phenotype of patients with CLN3 disease with the help of recent ophthalmic imaging modalities to distinguish CLN3 disease from other inherited retinal dystrophies.

METHODS

Patients underwent ophthalmic evaluations, including anterior and posterior segment examinations, optical coherence tomography, fundus autofluorescence, near infrared imaging and fundus photography. Patients were also assessed according to the Hamburg juvenile NCL (JNCL) score. Each ophthalmic finding was assessed by three independent examiners and assigned to a clinical severity score.

RESULTS

22 eyes of 11 patients were included. The mean age at examination was 14.4 years (range 11.8-26.4 years), with an average age at initial diagnosis of 8 years (range 4.5-11 years). The mean Hamburg JNCL score was 7.3 (range 0-13). All patients showed a specific macular striation pattern on optical coherence tomography that was independent of age and progression of the disease. Other previously described retinal features of CLN3 disease were classified into four severity grades.

CONCLUSIONS

This study represents the first prospective observational case series documenting retinal abnormalities in CLN3 disease with the aid of the spectral domain optical coherence tomography. The major finding was a characteristic, striated macular pattern in all patients studied. Particularly in early disease cases, macular striae can potentially help to discriminate CLN3 disease from other inherited forms of retinitis pigmentosa.

Multi-profile hidden Markov model for mood, dietary intake, and physical activity in an intervention study of childhood obesity

Motivated by an application to childhood obesity data in a clinical trial, this paper describes a multi-profile hidden Markov model (HMM) that uses several temporal chains of measures respectively related to psychosocial attributes, dietary intake, and energy expenditure behaviors of adolescents in a school setting. Using these psychological and behavioral profiles, the model delineates health states from the longitudinal data set. Furthermore, a two-level regression model that takes into account the clustering effects of students within school is used to assess the effects of school-based and community-based interventions and other risk factors on the transition between health states over time. The results from our study suggest that female students tend to decrease their physical activities despite a high level of anxiety about weight. The finding is consistent across intervention and control arms.

Study design and baseline description of the BMI2 trial: reducing paediatric obesity in primary care practices

OBJECTIVES

This study will test the efficacy of motivational interviewing (MI) conducted by primary care providers and dieticians among children ages 2-8 years old with a body mass index (BMI) ≥ 85th and ≤ 97th percentile.

METHODS

Forty-two practices from the American Academy of Pediatrics, Pediatric Research in Office Settings Network were assigned to one of three groups. Group 1 (usual care) measures BMI percentile at baseline, and at 1- and 2-year follow-ups and receives standard health education materials. Group 2 providers deliver three proactive MI counselling sessions with a parent of the index child in Year 1 and one additional 'booster' visit in Year 2. Group 3 adds six MI counselling sessions from a trained dietician. The primary outcome is the child's BMI percentile at 2-year follow-up. Secondary outcomes include parent report of the child's screen time, physical activity, intake of fruits and vegetables, and sugar-sweetened beverages.

RESULTS

We enrolled 633 eligible children whose mean BMI percentile was 92.0 and mean age of 5.1. The cohort was 57% female. Almost 70% of parents reported a household income of ≥ $40,000 per year, and 39% had at least a college education. The cohort was 63% white, 23% Hispanic, 7% black and 7% Asian. Parent self-reported confidence that their child will achieve a healthy weight was on average an 8 (out of 10).

CONCLUSION

To date, several aspects of the study can inform similar efforts including our ability to use volunteer clinicians to recruit participants and their willingness to dedicate their time, without pay, to receive training in MI.

Just enough, but not too much interactivity leads to better clinical skills performance after a computer assisted learning module

BACKGROUND

Well-designed computer-assisted instruction (CAI) can potentially transform medical education. Yet little is known about whether specific design features such as direct manipulation of the content yield meaningful gains in clinical learning. We designed three versions of a multimedia module on the abdominal exam incorporating different types of interactivity.

METHODS

As part of their physical diagnosis course, 162 second-year medical students were randomly assigned (1:1:1) to Watch, Click or Drag versions of the abdominal exam module. First, students' prior knowledge, spatial ability, and prior experience with abdominal exams were assessed. After using the module, students took a posttest; demonstrated the abdominal exam on a standardized patient; and wrote structured notes of their findings.

RESULTS

Data from 143 students were analyzed. Baseline measures showed no differences among groups regarding prior knowledge, experience, or spatial ability. Overall there was no difference in knowledge across groups. However, physical exam scores were significantly higher for students in the Click group.

CONCLUSIONS

A mid-range level of behavioral interactivity was associated with small to moderate improvements in performance of clinical skills. These improvements were likely mediated by enhanced engagement with the material, within the bounds of learners' cognitive capacity. These findings have implications for the design of CAI materials to teach procedural skills.

EGFR, TS, and ERCC1 expression in penile squamous cancer (PSC)

219

Background: PSC is an aggressive disease with devastating consequences. While near-universal EGFR overexpression has been documented, little is known about its clinical implications, and the potential role of KRAS mutations has not been described. In other tumor types, ERCC1 and TS expression are associated with response to platinum and 5FU chemotherapy, but have not yet been evaluated in PSC.

Methods: After IRB approval, 28 PSC patients treated at LAC-USC were identified with tumor specimens available. Testing for EGFR, TS, and ERCC1 expression relative to internal standard gene was performed by real-time RT-PCR and correlated with clinical features.

Results: The median age was 45 (30-78). There were 2 Black, 1 Asian, 1 White, and 15 Hispanic men; 9 had unspecified race. T stages included 4 Tis, 6 T1, 17 T2, and 9 T3; 11 men had lymph node (LN) involvement. No KRAS mutations were identified. EGFR had the highest relative expression (median 4.65, range 1.6-44.2), followed by TS (median 1.69, range 0.49-4.69); ERCC1 overexpression was rare (median 0.54, range 0.21-1.29). Higher EGFR expression was significantly associated with poor differentiation (median 12.5 compared to 3.6 for moderate/well differentiated tumors) on continuous (p=0.03 by Mann Whitney) and cut-point analysis using >7 (two- sided p=0.03 by Fisher's exact test) but did not correlate with stage. Stage > T2 showed a trend toward higher risk of LN involvement (p=0.06 by chi square). There was no correlation between differentiation and stage, and no significant correlation for ERCC1 or TS with grade or stage.

Conclusions: EGFR overexpression is common in PSC and correlates with tumor grade but not stage, suggesting it may be important for disease progression. The absence of KRAS mutations may portend responsiveness to EGFR- targeted therapy based on experience in other tumors. Low ERCC1 as an association with platinum response will be explored in an expanded cohort.

No significant financial relationships to disclose.

Final results of PX-171–003-A0, part 1 of an open-label, single-arm, phase II study of carfilzomib (CFZ) in patients (pts) with relapsed and refractory multiple myeloma (MM)

8504

Background: CFZ is a novel proteasome inhibitor of the epoxyketone class that exhibits a high level of proteasome selectivityand demonstrates antitumor activity in bortezomib (BTZ)-resistant MM pts in phase I studies. Methods: PX-171–003-A0 was an open-label, multicenter study that enrolled MM pts who relapsed from >2 prior therapies, failed BTZ and at least 1 immunomodulatory agent [thalidomide (THAL) or lenalidomide (LEN)], and were refractory to last treatment [progressing on or within 60 d of last therapy or <25% response to last therapy]. Pts received CFZ 20 mg/m2 IV d 1, 2, 8, 9, 15 and 16 every 28 d for up to 12 cycles (C). Clinical benefit response (CBR) was defined as MR or better. Results: 46 pts were enrolled, including 78% with progression on or within 60 d of last therapy and 22% with no response to last therapy. 39 pts completed at least 1 C of CFZ, had measurable M-protein, and were evaluable for response. Median prior therapies was 5 (range 2–15). 100% of pts received prior BTZ, 91% prior THAL, 89% prior LEN, and 83% prior stem cell transplant (SCT) and all had failed combinations including anthracyclines (80%) and/or alkylating agents (94%). Pts received a median of 3 C (range 1–12); 13 pts completed ≥6 C. CBR was 26% (10/39 eval pts), including 5 pts achieving PR and 5 pts achieving MR. 5 BTZ-refractory pts achieved MR or PR. Median TTP was 6.2 mo, the median DOR for the MR + PR was 7.4 mo. 8/10 pts achieved response during C1. 16 additional pts achieved SD for at least 6 wks. The most common adverse events were fatigue, anemia, thrombocytopenia, nausea, upper respiratory infection, increased creatinine and diarrhea. Peripheral neuropathy occured in < 10% of pts with 1 Gr 3 in a pt with pre-existing Gr 2. Conclusions: Single-agent CFZ achieved a TTP of > 6 mo in relapsed and refractory MM pts who failed available therapies. 26% of patients had at least an MR and median duration of >7 mo with this steroid- and anthracycline-sparing regimen. CFZ toxicities were manageable and importantly, exacerbation of pre-existing PN was rare. The study has been expanded to enroll an additional 250 pts in this unmet medical need population at an escalated dose, and treatment has been extended beyond a year.

[Table: see text]

Electronic band structure of Cu(2)O by spin density functional theory

The band structure of Cu(2)O is calculated using density functional theory in the generalized gradient approximation. By taking spin-orbit coupling into account the split between the Γ(7)(+) and the Γ(8)(+) valence band states is obtained as 128 meV. The highest valence band shows a noticeable nonparabolicity close to the Γ point. This is important for the quantitative description of excitons in this material, which is considered to be the best candidate for the confirmation that Bose-Einstein condensation also occurs in excitonic systems.

Aspirin responsiveness in acute brain ischaemia: association with stroke severity and clinical outcome

PURPOSE

Platelets play a critical role in the pathogenesis of acute brain ischaemia. We studied the association between the degree of inhibition of platelet function by aspirin (ASA) and the severity and outcome of acute brain ischaemia.

METHODS

Platelet responsiveness to ASA was assessed in patients with acute brain ischaemia, treated with ASA since hospital admission. The degree of ASA responsiveness was assessed by optical aggregometry and categorized into patients with good response, partial response and complete unresponsiveness to ASA (good responders, partial responders and non-responders, respectively). An additional evaluation of responsiveness to ASA was performed by Impact-R (cone and platelet analyzer). Patients underwent serial clinical assessment during hospitalization, at discharge and during follow-up.

RESULTS

Among 105 patients (mean age 63 +/- 12 years; 66% men), impaired ASA responsiveness at baseline as assessed by aggregometry was associated with increased stroke severity at baseline, unfavourable clinical course, and poor functional outcome during follow-up (p < 0.05 for all). Age-adjusted odds ratios in non-responders compared to good responders were 9.8 for severe stroke on admission (95% CI 2.8-34.9), 3.1 for lack of early clinical improvement (95% CI 1.1-8.8) and 8.6 for poor functional outcome during follow-up (95% CI 2.4-30.4). Less robust trends were observed with the Impact-R.

CONCLUSIONS

Impaired responsiveness to ASA in acute brain ischaemia is common and is associated with worse neurological deficits at stroke onset, early clinical deterioration and poorer functional outcome. The clinical significance of these findings requires further evaluation in larger longitudinal studies.

Modeling molecular interactions to understand spatial crowding effects on heterodimer formations

Molecular crowding occurs when the density of interacting molecules in some reaction system is sufficient to create deviations from traditional mass-action models of chemistry in diffusive systems. While there is a great deal of theory on the influence of molecular crowding on biochemistry in vivo, the effects are highly dependent on specific assumptions about the shapes, volumes, and diffusion properties of the components of an individual system and are thus difficult to predict from first principles. In this study, we use lattice Monte Carlo simulations to examine the effects on a reaction system for two limiting cases of the diffusion behavior of inert crowding agents. In cells, inert molecules might diffuse throughout a solute along with the reactant species by passive diffusion or may be anchored at fixed positions within the solute. We investigate the relative contributions of the two models to crowding effects by examining moving inert particles versus stationery inert particles on the kinetics of a heterodimer assembly system. The two models of inert crowding agents resulted in highly divergent effects on the reactant system. Stationary particles exhibited a bimodal response in the reaction rate curve that was a function of copy number and spatial arrangement and which accelerated the process at conditions not unlike those found in cellular environments. On the other hand, moving inert particles created a well mixed background that had no effect on the reaction process even under extremely compacted conditions. These results may have applications in developing more realistic simulations of reaction chemistry in crowded environments such as living cells.

Dioxins and dioxin-like PCBs in different fish from the river Elbe and its tributaries, Germany

In a long-term program polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/Fs) as well as dioxin-like polychlorinated biphenyls (DL-PCBs) were analyzed in the muscle tissue of eels (Anguilla anguilla), bream (Abramis brama), European chub (Leuciscus cephalus) and ide (Leuciscus idus) from the river Elbe and its tributaries Mulde and Saale. The variation of the PCDD/F and DL-PCB concentrations in all fish samples is very large, whereby the DL-PCBs predominate in comparison to the PCDD/Fs. In the eels, the concentrations (pg WHO-TEQ/g ww) for the PCDD/Fs lie in the range of 0.48-22 and for the DL-PCBs between 8.5 and 59. In the whitefish, the concentration range is 0.48-12 for the PCDD/Fs and 1.2-14 for the DL-PCBs. Statistical analysis using relative congener patterns for PCDD/Fs allow spatial correlations to be examined for sub-populations of eels and whitefish. The results are compared to the maximum levels laid down in the European Commission Regulation (EC) No. 466/2001 and the action levels of the European Commission Recommendation 2006/88/EC. Eels caught directly after the major flood in August 2002 as well as eels near Hamburg (years 1996 and 1998) show high concentration peaks. Compared to the eels whitefish is less contaminated with PCDD/Fs and DL-PCBs.

HER2 expression as a potential marker for response to therapy targeted to the EGFR

Since human epidermal growth factor receptor 2 (HER2) is known to participate with the epidermal growth factor receptor (EGFR) in mitogenic signalling, we hypothesised that HER2 overexpression might indicate responsiveness to EGFR targeted therapies. MCF7 breast cancer cells transfected with the HER2 gene were subcloned to establish a set of genetically related cell lines expressing graded levels of HER2 by immunoblot analysis. The subcloned cell lines and parental MCF7 cells were characterised by their growth characteristics, and cell by cell patterns of EGFR, HER2 and HER3 expression as well as levels of phosphorylated mitogen-activated protein kinase (MAPK) and AKT by laser scanning cytometry (LSC). Growth inhibition assays were used to characterise response to EGFR targeted therapy, and to determine the relationship between therapeutic response and levels of tyrosine kinase expression. The levels of growth inhibition of AG1478 and of the AG1478-trastuzumab combinations were correlated with levels of HER2 expression among the different cell lines. Among EGFR, HER2 and HER3, HER2 overexpression was the best single predictive marker, but combinations of two markers provided additional predictive information.

Contamination of the alluvial plain, feeding-stuffs and foodstuffs with polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans (PCDD/Fs), dioxin-like polychlorinated biphenyls (DL-PCBs) and mercury from the River Elbe in the light of the flood event in August 2002

Meadow soils, feeding-stuffs and foodstuffs from the alluvial plain of the river Elbe were analyzed in respect of PCDD/Fs, DL-PCBs and mercury with a view to assessing the consequences of the extreme flood of August 2002. The PCDD/F concentrations in the soils range from 3 to 2100 ng WHO-TEQ/kg dm, and for the DL-PCBs the range was 0.32 to 28 ng WHO-TEQ/kg dm. On the basis of established threshold values >40% of the areas are only fit for restricted usage. Mercury concentrations range from 0.11 to 17 mg/kg dm, whereby the action value of 2 mg/kg dm is exceeded in about 50% of the soil samples. A cumulative memory effect from past floods rather than a recent contamination from August 2002 is documented. Soils taken from behind broken dykes showed significantly lower concentrations. Grass, hay and grass silage originating from pasture land in Lower Saxony were taken before and immediately after the flooding. PCDD/Fs range from 0.29 to 16 ng WHO-TEQ/kg, the maximum permitted value of 0.75 ng WHO-TEQ/kg was exceeded in about 50% of the samples. Muscle-tissue from cattle, sheep, lamb and a roe deer as well as untreated milk from individual cows returned values ranging from 0.76 to 5.9 pg WHO-PCDD/F-TEQ/g fat, and 10% of the samples returned values higher than the permitted maximum of 3 pg WHO-PCDD/F-TEQ/g fat. The action value of 2 pg WHO-PCDD/F-TEQ/g fat was exceeded in 33% of the samples. No direct connection between these results and the effects of the flood could be established. A major input path for PCDD/Fs is the tributary Mulde, which discharges contaminated sediments from its catchment area into the Elbe.

Symptom assessment tools used in a hospital-based pain and symptom care program at the University of North Carolina–A novel proxy rating system

8579

Background: Level of consciousness (LOC) often varies in the palliative care patient population. Tools are needed to reliably rate symptoms. The purpose of our analysis was to describe the patient demographics and rating distributions elicited by two distinct measurement tools, one for those able to self-rate (SR) and one for proxy-rated (PR) patients. Methods: Data obtained prospectively over the time period 1/2002–12/2004 was reviewed. A symptom distress scale using 11 items on a 5 point Likert-type scale (5=most severe) was adapted from the revised McCorkle symptom distress scale and used for alert and oriented patients. Symptoms evaluated included nausea, mood, appetite, shortness of breath (SOB), insomnia, pain, mobility, fatigue, constipation, decreased LOC, and delirium. For patients unable to SR, an advanced practice nurse rated symptoms using the Detroit Medical Center non-verbal pain scale and proxy symptom distress measures adapted from the Common Toxicity Criteria Manual, version 2. Rated symptoms included pain, vomiting, constipation, tachypnea, depressed LOC, and agitated delirium. Results: A total of 259 patients with data from visits 1 and 2 were included in the analysis. On visit 1, 46% (61/133) of PR patients were male, 70% (94/134) were ≥60yrs and 73% (98/134) were Caucasian. Symptom distress scores varied by rater and visit (Table). Conclusions: There is a need for proxy symptom distress measurement in the palliative care patient population as LOC fluctuates. Variability in symptom distress ratings between the SR and PR groups, with a trend toward lower scores in PR patients, is in accordance with the current literature. Of note, there was a consistent improvement in scores, regardless of rater, when visit 2 was compared to visit 1. We plan a prospective study of inter-rater reliability and validity of this novel symptom scoring system.

[Table: see text]

No significant financial relationships to disclose.

Expectation-maximization method for reconstructing tumor phylogenies from single-cell data

Recent studies of gene expression in cancerous tumors have revealed that cancers presenting indistinguishable symptoms in the clinic can represent substantially different entities at the molecular level. The ability to distinguish between these different cancers makes possible more accurate prognoses and more finely targeted therapeutics. Making full use of this knowledge, however, requires characterizing commonly occurring cancer sub-types and the specific molecular abnormalities that produce them. Computational approaches to this problem to date have been hindered by the fact that tumors are highly heterogeneous masses typically containing cells at multiple stages of progression from healthy to aggressively malignant. We present a computational approach for taking advantage of tumor heterogeneity when characterizing tumor progression pathways by inferring those pathways from single-cell assays. Our approach uses phylogenetic algorithms to infer likely evolutionary sequences producing cell populations in single tumors, which are in turn used to create a profile of commonly used pathways across the patient population. This approach is combined with expectation maximization to infer unknown parameters used in the phylogeny construction. We demonstrate the approach on a set of fluorescent in situ hybridization (FISH) data measuring cell-by-cell gene and chromosome copy numbers in a large sample of breast cancers. The results validate the proposed computational methods by showing consistency with several previous findings on these cancers. They also provide novel insights into the mechanisms of tumor progression in these patients.