NOTE: Extended to 5/31/2009 (from 11/30/2008) per PI (7/09)

NOTE: Received NCE to 11/30/2008 per JSC (8/08)

NOTE: Received NCE to 8/31/2008 per JSC (11/07)

NOTE: End date changed back to 01/27/2008 per S. Krenek/JSC (8/07)

End date changed to 3/31/2008 per JSC info update (10/06)

The study of risk factors focuses on the types and doses of radiation exposure in space flight, on measures of terrestrial solar UV radiation exposure, measures of nutrition, smoking, and general health. The assessment of cataract will use standardized digital images of the lens and validated measures of severity. The study will measure the severity and progression rates of cortical, nuclear, posterior subcapsular, and mixed cataracts, and it will relate these measures to measures of UV and ionizing radiation, nutrition, smoking, and general health. A secondary goal of this project will be to improve the routine annual clinical assessment of the ocular lens by including Nidek EAS 1000 digital imaging of the lens in the annual ocular examination protocol for astronauts.

NASCA contains an initial cross-sectional assessment of the severity of the three types of lens opacification, and a follow-on assessment of progression rates of the three classes lens opacification in the populations of astronauts, military aircrew, and ground based comparison participants in the Longitudinal Study of Astronaut Health (LSAH).

The main goal of the five-year longitudinal study is to determine the progression rates of the three main types of lens opacification in astronaut who have flown at least one mission in space, and a comparison group made up of astronauts who have not yet flown in space and military aircrew members. We will then determine the risk factors associated with cataract progression with a specific focus on the components and doses of radiation exposure during space flight. Specifically, total radiation lens dose, space radiation lens dose, and individual contributions from space galactic cosmic ray and trapped proton lens dose will be assessed.

NASA is also concerned about identifying and ultimately mitigating the risks to astronaut health of exposure to radiation in space. Several avenues of research now suggest that increased risk of lens opacification may be one of these adverse health effects. In order to address this issue, NASA has approved and funded this five-year, multi-centered research proposal entitled “The Precise Assessment of Prevalence and Progression of Lens Opacities in Astronauts as a Function of Radiation Exposure During Space Flight.”

The Brigham and Women’s Hospital (BWH), the Johnson Space Center (JSC), the Departments of Medicine at Baylor College of Medicine (BCM) and The Methodist Hospital (TMH) both in Houston, Wyle Laboratories, and Space Center Eye Associates are the six centers cooperating in the execution of this study. The epidemiologic team at BCM and TMH will recruit astronauts and control subjects. Members of the Flight Medicine Clinic at the JSC and will perform ocular and general medical examinations. They will obtain specialized digital images of the crystalline ocular lens that will enable investigators at The Center for Ophthalmic Research (COR) at the BWH to derive measures of the severity for each class of cataract. Wyle Laboratories personnel at JSC will create and maintain the main data set of this project, and members of the Radiation Safety Office and the Statistical Branch of the JSC will work with Dr. Chylack, the PI of this project, and the other Co-Investigators to analyze of the data from this project and prepare regular reports and manuscripts.

Leo T. Chylack, Jr., M.D. (BWH) originally submitted the proposal as a Supplemental Medical Objective (SMO). The goal of the proposal was to supplement the current annual ocular examination with new measurements that would allow an objective and more quantitative routine assessment of the status of the crystalline lens in astronauts. This supplementary methodology has enabled NASA to obtain objective assessments of the clarity of the lens, quantitative, continuous measures of the severity of lens opacification, and, over time, cataract type-specific progression rates.

The SMO has undergone review by several intramural boards and a non-advocate peer review (NAR) panel. As a result of this review, the NAR recommended 1) a major expansion of the project to enable measures of severity and progression rates of various forms of lens opacification, and 2) modifications of the number and composition of the control groups to provide comparisons of severity and progressions rates of lens opacification in astronauts to those in military exposed to the cataractogenic risks of high altitude (but not space) flight and to those in LSAH subjects, who presumably are not exposed to high altitude or space radiation.

LIMITATIONS OF THE NASCA STUDY: 1) relatively small sizes of cohorts studied, 2) no perfect comparison group (astronaut who flew in space w/o radiation exposure), 3) combining the two comparison groups (astronauts who did not fly in space + military aircrew), albeit reasonable, was not ideal, 4) NASA’s group of ground-based subjects, each of whom was matched to a newly selected astronaut and designated a “control” subject for that astronaut, proved to be very dissimilar to the astronaut group in the NASCA study and had to be dropped from the cross-sectional and longitudinal 5-year data analyses, 5) space radiation exposures were generally on the low side, so the utility of individual dose data was limited, 6) precision of measured doses of space radiation exposure was also limited, 7) distributions of primary end-point variables were non-normal and often skewed requiring highly specialized statistical methods.

STRENGTHS OF THE NASCA STUDY: 1) well-defined population of astronauts, 2) highly standardized methods used for all measures, 3) sensitive, precise, continuous, and validated measures of cataract severity, 4) these methods clearly detected the age-related changes in each class of lens opacification. Opacities increased slowly with age, as expected, 5)Excellent consistency and concordance of results for cortical opacification among the cross-sectional and longitudinal studies, 6) moderate consistency and concordance of results for PSC opacification among the cross-sectional and longitudinal studies, 7) results of the analyses relating nuclear cataract to space radiation exposure less consistent, 8) the lack of a dose-response relationship between dose of space radiation and rate of progression of lens opacification may be due in part to non-targeted effects (Cucinotta & Chappell, Mutation Res., 2010) in which the RBE at low doses is much higher than previously expected.

IMPLICATIONS OF THE NASCA STUDY: 1) Adverse effect of space radiation exposure on the lens (particularly on cortical and PSC) are evident in a five year follow-up study, 2) the adverse effects are greater for cortical and PSC than nuclear opacification, although the cross-sectional data showed an effect on nuclear as well, 3) further study of the relationship between space radiation exposure is needed, but the study will be complicated by the need for longer-term follow-up (>5 years), the lack of ideal or even optimal comparison subjects, and relatively small numbers of astronauts available to study, 4) it may be preferable to study only astronauts who have flown in space and compare cohorts with higher-doses to those with lower-dose exposures, rather than use non-astronaut comparison subjects, 5) increased precision of assessing the doses space radiation exposure by radiation source is needed, 6) it is likely that worsening of lens opacification will manifest itself on prolonged manned space missions (e.g. to Mars), but NASCA data suggest that the consequences to high- and low-contrast visual acuity will be small.

The study is complete; we present our analyses of 5-6 years of data. The first section deals with CROSS-SECTIONAL STUDIES OF: 1) BASELINE DATA and 2) FIVE YEARS-WORTH OF DATA.

BASELINE CROSS-SECTIONAL ANALYSIS:

In our publication of baseline analyses (Chylack LT Jr. et al. NASA Study of Cataract in Astronauts (NASCA). Report 1: Cross-Sectional study of the relationship of exposure to space radiation and risk of lens opacity. Rad. Res. 2009;172:10-20.), we analyzed baseline data from 171 astronauts who flew at least one mission, 53 astronauts who had not flown in space, 95 military aircrew, and 99 non-aircrew ground-based comparison subjects.

We used continuous measures of severity of lens opacification (C, N, and PSC) that were derived from images obtained with the Nidek EAS 1000 Lens Imaging System. Astronauts who flew at least one mission were matched to comparison subjects with Propensity Scores based on demographic characteristics, medical history stratified by gender and smoking (ever/never). Customized non-normal regression models examining effect of space radiation on each measure of opacity. We found that 1) the median severity and variability in the size of cortical opacities were greater in exposed astronauts (p=0.015), 2) galactic cosmic rays (GCR) may be linked to increased PSC area (p=0.056) and the number of PSC centers (p=0.095), 3) within the astronaut group PSC severity (area) was greater in subjects with higher radiation doses (p=0.016).

CROSS-SECTIONAL ANALYSIS OF FIVE YEARS-WORTH OF DATA:

After completing the five-year study we examined the relationship between cataract severity and space radiation exsposure using a non-parametric, cross-sectional analysis using all five years-worth of data. We used the the median maximum opacity (OD,OS) over all visits as primary outcome variable, and customized regression models to make inferences on possible space radiation effects on severity of C, N, and PSC, also adjusting for confounding variables. Each regression model was chosen to most closely accommodate the distribution of the outcome variable (opacity type) to be studied.

The results of this analysis are as follows by cataract class: (Nuclear, Cortical, and Posterior Subcapsular). For Nuclear we used as measures of cataract severity: 1) the ave. pixel density within oval mask of entire nucleus, 2) the average pixel density in smaller masks in the central clear zone (CCZ) and the anterior embryonal nucleus (AEN) of the nucleus. All measures were statistically significantly (p=0.02) linked to exposure to space radiation (yes/no). For Cortical we used as a measure of severity the ave. % area opaque in the digital, binarized image of the lens. The variance (pleomorphic character of the opacity) was linked directly with exposure to space radiation (p=0.000) but the mean severity of cataract was not linked to exposure to space radiation (p=NS). For posterior subcapsular cataract (PSC) we used the average % area opaque in a digital binarized image and also the # of centers of PSC. Neither of these measures of PSC was linked to exposure to space radiation.

The conclusions from the five-year cross-sectional analysis were: 1) we confirmed the findings of the baseline cross-sectional analysis that there is a direct relationship between the variance in severity, but not the mean severity, of cortical cataract and exposure to space radiation, and 2) the average densities of nuclear opacification in the CCZ, AEN, and entire oval mask were significantly and directly related to exposure to space radiation.

LONGITUDINAL ANALYSIS OF NASCA DATA:

In these studies we were examining the relationship of exposure to space radiation and the rates of cataract progression for Nuclear, Cortical and PSC.

Methods: All subjects had eye exams at NASA. Digital lens images (Nidek EAS 1000) were obtained. Nutritional data were derived from standardized questionnaires. Because of high variability and skewness of opacity measures, nonparametric methods were used to test for association between rates of opacification and space radiation exposure. First, median regression was used to obtain robust estimates of slopes of opacity vs. time for each eye of each subject. Next, a partial correlation analogue of Kendall's Tau was used to quantify and test radiation effect on slopes, adjusting for confounding variables age, nutritional and sun-exposure histories, with std. errors adjusted for repeated observations on each eye for each subject. The dose of exposure to space radiation of the lens was estimated using recorded personal dosimetry from each space mission, adjusted by radiation quality factor, and quantified in terms of dose equivalent and dose equivalent latency for total & galactic cosmic radiation. For Ast0Ex and Mil, radiation exposure was set = zero. In addition to the analysis of growth rates, we similarly analyzed median values of max (OD,OS) opacity over the multiple visits for each subject. In these longitudinal analysis of the relationship between cataract progression rates and exposure the space radiation we observed the following trends: 1) estimates of progression rates exhibited high variability and skewed distributions, and 2) regression models assuming normally distributed errors were deemed inappropriate. Consequently, we used a two-step approach: as a first step we used median regression to estimate slope (opacity change/yr.); and as a second step we used PC-tau, a partial correlation analog of Kendall’s Tau, to quantify association between opacity growth and various measures of space radiation. We adjusted for age, sun exposure, nutrition and other confounding factors. PC-taus were calculated considering all possible between-subject comparisons, keeping in mind that most subjects had two opacity slopes (OD and OS), and then inferences on the effect of radiation were then made on the basis of PC-Tau, relative to its standard error.

RESULTS OF LONGITUDINAL STUDY:

(Visual Acuity): In these partial tau analyses the effects of age, pilots, two nutritional variables, and smoking have been removed. Obese astronauts removed. We found no apparent relationship between space radiation exposure and change in high or low contrast measures of visual acuity.

(Cataract): The major conclusions are as follows: 1) Space radiation exposure (yes/no) has a statistically significant relationships with faster rates of progression of cortical opacification (p=0.02) and increased numbers of loci of posterior subcapsular opacification (p=0.04), 2) Space radiation exposure (yes/no) has no statistically significant relationships with faster rates of progression of either nuclear or posterior subcapsular opacification.

SUMMARY OF NASCA FINDINGS from the Cross-Sectional Study (severity of cataract vs. space radiation exposure), and the Longitudinal Study (progression rate of cataract vs. space radiation exposure):

FINDINGS OF CROSS-SECTIONAL BASELINE STUDY: 1) The median size and variance of cortical opacities were greater in exposed astronauts (p=0.015), 2) Within astronaut group PSC severity (area) was greater in subjects with higher radiation doses (p=0.016). Galactic cosmic rays (GCR) may be linked to increased PSC area (p=0.056) and the number of PSC centers (p=0.095), 3) No relationship between severity of nuclear opacification and space radiation exposure.

FINDINGS OF CROSS-SECTIONAL STUDY OF 5-YEARS-WORTH OF DATA: 1) The variance in size, but not the mean size, of cortical opacities was greater in exposed astronauts (p=0.000), 2) Neither size nor number of centers of PSC was linked to space radiation exposure (yes/no), 3) The average density of nuclear opacification in the CCZ, AEN, and entire oval mask was significantly related to exposure to space radiation (yes/no) (p=0.02).

FINDINGS OF LONGITUDINAL STUDY OF 5-YEARS-WORTH OF DATA: 1) The rate of progression of cortical opacification was significantly higher in exposed astronauts (p=0.02), 2) The increase in numbers of centers, but not the change in area, of posterior subcapsular opacification was significantly higher (p=0.04) in astronauts exposed to space radiation (yes/no), 3) Space radiation exposure (yes/no) had no statistically significant relationships with faster rates of nuclear opacification.

LIMITATIONS OF THE NASCA STUDY: 1) relatively small sizes of cohorts studied, 2) no perfect comparison group (astronaut who flew in space w/o radiation exposure), 3) combining the two comparison groups (astronauts who did not fly in space + military aircrew), albeit reasonable, was not ideal, 4) NASA’s group of ground-based subjects, each of whom was matched to a newly selected astronaut and designated a “control” subject for that astronaut, proved to be very dissimilar to the astronaut group in the NASCA study and had to be dropped from the cross-sectional and longitudinal 5-year data analyses, 5) space radiation exposures were generally on the low side, so the utility of individual dose data was limited, 6) precision of measured doses of space radiation exposure was also limited, 7) distributions of primary end-point variables were non-normal and often skewed requiring highly specialized statistical methods.

STRENGTHS OF THE NASCA STUDY: 1) well-defined population of astronauts, 2) highly standardized methods used for all measures, 3) sensitive, precise, continuous, and validated measures of cataract severity, 4) these methods clearly detected the age-related changes in each class of lens opacification. Opacities increased slowly with age, as expected, 5) excellent consistency and concordance of results for cortical opacification among the cross-sectional and longitudinal studies, 6) moderate consistency and concordance of results for PSC opacification among the cross-sectional and longitudinal studies, 7) results of the analyses relating nuclear cataract to space radiation exposure less consistent, 8) the lack of a dose-response relationship between dose of space radiation and rate of progression of lens opacification may be due in part to non-targeted effects (Cucinotta & Chappell, Mutation Res., 2010) in which the RBE at low doses is much higher than previously expected.

IMPLICATIONS OF THE NASCA STUDY: 1) Adverse effect of space radiation exposure on the lens (particularly on cortical and PSC) are evident in a five year follow-up study, 2) the adverse effects are greater for cortical and PSC than nuclear opacification, although the cross-sectional data showed an effect on nuclear as well, 3) further study of the relationship between space radiation exposure is needed, but the study will be complicated by the need for longer-term follow-up (>5 years), the lack of ideal or even optimal comparison subjects, and relatively small numbers of astronauts available to study, 4) it may be preferable to study only astronauts who have flown in space and compare cohorts with higher-doses to those with lower-dose exposures, rather than use non-astronaut comparison subjects, 5) increased precision of assessing the doses space radiation exposure by radiation source is needed, 6) it is likely that worsening of lens opacification will manifest itself on prolonged manned space missions (e.g. to Mars), but NASCA data suggest that the consequences to high- and low-contrast visual acuity will be small.

NOTE: Extended to 5/31/2009 (from 11/30/2008) per PI (7/09)

NOTE: Received NCE to 11/30/2008 per JSC (8/08)

NOTE: Received NCE to 8/31/2008 per JSC (11/07)

NOTE: End date changed back to 01/27/2008 per S. Krenek/JSC (8/07)

End date changed to 3/31/2008 per JSC info update (10/06)

NASCA contains an initial cross-sectional assessment of the severity of the three types of lens opacification, and a follow-on assessment of progression rates of the three classes lens opacification in the populations of astronauts, military aircrew, and ground based comparison participants in the Longitudinal Study of Astronaut Health (LSAH).

The main goal of the five-year longitudinal study is to determine the progression rates of the three main types of lens opacification in the complete sample of astronauts, the control populations of military aircrew and ground-based subjects in the LSAH, and then determine the risk factors associated with cataract progression with a specific focus on the components and doses of radiation exposure during space flight. Specifically, total radiation lens dose, space radiation lens dose, and individual contributions from space galactic cosmic ray and trapped proton lens dose will be assessed.

NASA is also concerned about identifying and ultimately mitigating the risks to astronaut health of exposure to radiation in space. Several avenues of research now suggest that increased risk of lens opacification may be one of these adverse health effects. In order to address this issue, NASA has approved and funded this five-year, multi-centered research proposal entitled “The Precise Assessment of Prevalence and Progression of Lens Opacities in Astronauts as a Function of Radiation Exposure During Space Flight.” The Brigham and Women’s Hospital (BWH), the Johnson Space Center (JSC), the Departments of Medicine at Baylor College of Medicine (BCM) and The Methodist Hospital (TMH) both in Houston, Wyle Laboratories, and Space Center Eye Associates are the six centers cooperating in the execution of this study. The epidemiologic team at BCM and TMH will recruit astronauts and control subjects. Members of the Flight Medicine Clinic at the JSC and will perform ocular and general medical examinations. They will obtain specialized digital images of the crystalline ocular lens that will enable investigators at The Center for Ophthalmic Research (COR) at the BWH to derive measures of the severity for each class of cataract. Wyle Laboratories personnel at JSC will create and maintain the main data set of this project, and members of the Radiation Safety Office and the Statistical Branch of the JSC will work with Dr. Chylack, the PI of this project, and the other Co-Investigators to analyze of the data from this project and prepare regular reports and manuscripts.

Leo T. Chylack, Jr., M.D. (BWH) originally submitted the proposal as a Supplemental Medical Objective (SMO). The goal of the proposal was to supplement the current annual ocular examination with new measurements that would allow an objective and more quantitative routine assessment of the status of the crystalline lens in astronauts. This supplementary methodology has enabled NASA to obtain objective assessments of the clarity of the lens, quantitative, continuous measures of the severity of lens opacification, and, over time, cataract type-specific progression rates.

The SMO has undergone review by several intramural boards and a non-advocate peer review (NAR) panel. As a result of this review, the NAR recommended 1) a major expansion of the project to enable measures of severity and progression rates of various forms of lens opacification, and 2) modifications of the number and composition of the control groups to provide comparisons of severity and progressions rates of lens opacification in astronauts to those in military exposed to the cataractogenic risks of high altitude (but not space) flight and to those in LSAH subjects, who presumably are not exposed to high altitude or space radiation.

Since January 2007, the project has proceeded apace. There have been no significant delays in seeing subjects, and we anticipate that our three cohorts will complete two full years of follow-up between December 2007 and May, 2008. The last ground-based control subject completed two full years of follow-up on December 14, 2007; the last military aircrew will complete two years of follow-up on March 16, 2008, and the last astronaut to complete two full years of follow-up will be seen on or about May 25, 2008. I emphasize this two-year follow-up interval, because the Non-Advocate Review (NAR) Panel which was convened in April, 2007 recommended that we assess the longitudinal data from NASCA after all subjects had completed two full years of follow-up. This recommendation originated in discussions of the need to ascertain the existence of a longitudinal signal and the need to estimate the power of the study to detect a relationship in five years between space radiation exposure and lens opacity progression rates.

Because we will not be able to complete a longitudinal analysis of the complete two-year data set until late fall 2008, we have decided to conduct preliminary statistical analysis of completed longitudinal data sets in February 2008. We hope to identify any longitudinal trends relating the dose of space radiation exposure to progression rates of C, N, and PSC opacification. The complete NASCA data set will not be proofread and analyzed before the end of the year. We hope that the preliminary analysis, if positive, will allow us to seek extended funding for the project without losing momentum. If the preliminary analyses are inconclusive, we will have to wait until the fall to conduct an analysis of the entire two-year data set. As explained above, it is not clear how we are going to proceed beyond the end of the NCE without funding. Notwithstanding the above, we are on track to complete these tasks on this timeline.

There has been an important change in the manner in which we are approaching the statistical analyses of the data. In our report last year we compared astronauts to controls, regardless of whether or not the astronaut had flown in space. We were concerned that by including astronauts who had not flown in space, we might be reducing our chance to see the relationships we were investigating. Now we are using propensity scores to match astronauts who have flown in space to those controls without space flight experience (astronauts who have not flown, military aircrew, and ground-based controls). Exposed astronauts were matched to an equal number of non-exposed control subjects using a propensity score based on demographic characteristics and medical history, stratified by gender and whether or not the subject ever smoked. For each stratum, the propensity score was a linear combination of the five best predictors of exposure with coefficients estimated from a logistic regression model. Using the body mass index (BMI) in the propensity model, even with height and weight as separate terms, significantly improved the overall fit in characterizing exposed subjects. This is possible because BMI is a nonlinear function of height and weight. Age was not included as a propensity score variable, because exposed astronauts tended to be older than non-exposed astronauts, and we considered it important that all astronauts be in the matched sample. Adjustments for age differences were made in the regression models. For each of the four strata defined by gender and smoking history, each of the 171 exposed astronauts were then matched with the available (i.e. not already matched) subject; having the closest possible propensity score. An exception to this rule was made to ensure that all 53 members of the astronaut corps who hadn't yet flown were matched with an exposed astronaut.

We have also modified out analysis methodologies: opacity data were highly skewed and/or were partially discrete, which obviated use of models with normally distributed errors for making inference on the effect of space radiation on lens opacification. Instead, we used customized non-normal regression models with parameters (or transformed parameters) modeled as linear combinations of explanatory variables to test if exposure to space radiation was associated with higher N, C, or PSC opacification after adjustment for age, sun exposure, occupational group (astronaut, aircrew, or ground), baseline differences, and nutritional intake. More specifically, we used beta regression for scaled N mask density, skew-normal regression for log-transformed C opacity, and ordered logit regression for PSC opacity and counts of PSC centers. Dependent variables for all analyses were the maximum measure of opacity between the right (OD) and left (OS) eyes. Maximum N mask density was scaled from 0 to 1 (instead of the original pixel count ranging from 0 to 255), maximum C opacity was log-transformed to fit the skew-normal model, while maximum PSC opacity (% opaque) , because of its discrete character, large quantities of zeros and extreme skewness, was grouped into five bins: #1 (0) #2 (.01 - .03), #3 (0.04 – 0.06), #4 (0.07 – 0.16) and #5 (> 0.16). These bins were chosen to make the overall proportion of observations in each bin as close as possible to each other. PSC center data was in the form of total counts of opacity islands per eye. Analyses were run only on data from exposed and matched subjects. Remaining subjects were considered too unlike the exposed group, and were thus not included in these cross-sectional analyses because of potential for biasing the results. However some or all of these subjects could be included in future analyses of longitudinal data.

A secondary objective of these analyses was to identify nutritional covariates that appeared to ameliorate or exacerbate opacification. Therefore, in the beta regression model for nuclear opacity, alpha-carotene, beta carotene, lutein + zeaxanthin, and poly-unsaturated fats were included as possible explanatory variables. In the skew-normal regression model for C opacity, alpha-carotene, other carotenoids, and vitamin A were incorporated. Lastly, in the ordered logit regression model for PSC opacity, only lycopene was incorporated with other covariates.

In January 2008 we submitted to Investigative Ophthalmology and Visual Science (IOVS) a manuscript with results of an analysis of our baseline data. We found that the variability and median area of C cataract were significantly higher for exposed astronauts than for a hypothetical group of non-exposed astronauts with similar ages (P = 0.015), and galactic-cosmic space radiation may be linked to increased PSC area (P = 0.056) and the number of PSC centers (P = 0.095). No association was found between space radiation and increased N. We concluded that the cross-sectional data analysis revealed a small deleterious effect of space radiation for C and possibly for PSC at smaller radiation doses than had been reported previously.

There were also statistically significant relationships between certain nutrients and lens opacification. Nutritional information collected in NASCA may also unravel uncertainties surrounding the role of various nutrients on the progression of opacities. Our results for N opacification suggest a protective effect of beta-carotene. Our findings also suggest a significant protective effect of lycopene intake on PSC density and centers.

NOTE: End date changed back to 01/27/2008 per S. Krenek/JSC (8/07)

End date changed to 3/31/2008 per JSC info update (10/06)

NASCA contains an initial cross-sectional assessment of the severity of the three types of lens opacification, and a follow-on assessment of progression rates of the three classes lens opacification in the populations of astronauts, military aircrew, and ground based comparison participants in the Longitudinal Study of Astronaut Health (LSAH).

The main goal of the five-year longitudinal study is to determine the progression rates of the three main types of lens opacification in the complete sample of astronauts, the control populations of military aircrew and ground-based subjects in the LSAH, and then determine the risk factors associated with cataract progression with a specific focus on the components and doses of radiation exposure during space flight. Specifically, total radiation lens dose, space radiation lens dose, and individual contributions from space galactic cosmic ray and trapped proton lens dose will be assessed.

NASA is also concerned about identifying and ultimately mitigating the risks to astronaut health of exposure to radiation in space. Several avenues of research now suggest that increased risk of lens opacification may be one of these adverse health effects. In order to address this issue, NASA has approved and funded this five-year, multi-centered research proposal entitled “The Precise Assessment of Prevalence and Progression of Lens Opacities in Astronauts as a Function of Radiation Exposure During Space Flight.” The Brigham and Women’s Hospital (BWH), the Johnson Space Center (JSC), the Departments of Medicine at Baylor College of Medicine (BCM) and The Methodist Hospital (TMH) both in Houston, Wyle Laboratories, and Space Center Eye Associates are the six centers cooperating in the execution of this study. The epidemiologic team at BCM and TMH will recruit astronauts and control subjects. Members of the Flight Medicine Clinic at the JSC and will perform ocular and general medical examinations. They will obtain specialized digital images of the crystalline ocular lens that will enable investigators at The Center for Ophthalmic Research (COR) at the BWH to derive measures of the severity for each class of cataract. Wyle Laboratories personnel at JSC will create and maintain the main data set of this project, and members of the Radiation Safety Office and the Statistical Branch of the JSC will work with Dr. Chylack, the PI of this project, and the other Co-Investigators to analyze of the data from this project and prepare regular reports and manuscripts.

Leo T. Chylack, Jr., M.D. (BWH) originally submitted the proposal as a Supplemental Medical Objective (SMO). The goal of the proposal was to supplement the current annual ocular examination with new measurements that would allow an objective and more quantitative routine assessment of the status of the crystalline lens in astronauts. This supplementary methodology has enabled NASA to obtain objective assessments of the clarity of the lens, quantitative, continuous measures of the severity of lens opacification, and, over time, cataract type-specific progression rates.

The SMO has undergone review by several intramural boards and a non-advocate peer review (NAR) panel. As a result of this review, the NAR recommended 1) a major expansion of the project to enable measures of severity and progression rates of various forms of lens opacification, and 2) modifications of the number and composition of the control groups to provide comparisons of severity and progressions rates of lens opacification in astronauts to those in military exposed to the cataractogenic risks of high altitude (but not space) flight and to those in LSAH subjects, who presumably are not exposed to high altitude or space radiation.

Recruitment for NASCA began in July 2004. Our goal was to recruit all of the 285 US astronauts, 100 military aircrew members, and 100 ground-based controls. Most of our subjects were enrolled wihtin the first 18 months of active recruiting, a few were added in the subsequent six months. As of June 2006, we had enrolled 224 astronauts (78.6% of goal), 95 military aircrew (95% of goal), and 99 ground-based controls (99% of goal). The shortfall in enrolled astronauts was due largely to death, intervening cataract surgery (an exclusion criterion), and disinclincation to participate among older astronauts. The NASCA project is proceeding with these three cohorts.

Over the summer of 2006 we completed the statistical analysis of the baseline data - the first year's data - and prepared two manuscripts: 1) describing the recruiting procedures and results, the methodologies used, the propensity scores for each cohort, and the variables likely to confound the statistical analysis, and 2) describing the association between space radiation exposure,solar UV radiation exposure, nutrition, smoking, general health factors, and medications used to the severity of cortical (C), nuclear (N), and posterior subcapsular (P) lens opacification.

In order to facilitate the abbreviation of these reports, we set up a second NASCA web site using a server that would hold all of the detailed materials about techniques, methods, policies, etc. and be accessible to the public. By removing all of the fine methodological detail from the two manuscripts and locating them on the publically accessible web site, we were able to meet the journal's requirements about manuscript length without limiting disclosure of the study's methods. The web site's URL is ( http://nasca-study.bwh.harvard.edu).

A summary of the findings of the two manuscripts is provided below:

MANUSCRIPT 1:

TITLE: NASCA Report 1: Study of Relationship of Exposure to Space Radiation and Risk of Lens Opacity

AUTHORS: Leo T. Chylack, Jr.(a), M.D., Leif E. Peterson, Ph.D.(b), Alan H. Feiveson, Ph.D.(c), Mary Wear, Ph.D.(e), F. Keith Manuel, O.D.(e), William H. Tung (a), Dale Hardy, M.S.(b), Lisa Marak, R.N.(d), and Francis A. Cucinotta, Ph.D.(c)

AFFILIATIONS: a) Center for Ophthalmic Research, Brigham and Women's Hospital, Boston, MA; b) Department of Medicine , Baylor College of Medicine, Houston, TX; c) Lyndon B. Johnson Space Center, NASA, Houston, TX; d) Wyle Laboratories, Houston, TX; e) Space Center Eye Associates, Houston, TX.

ABSTRACT:

Objectives: The NASA Study of Cataract in Astronauts (NASCA) is a five-year investigation of lens opacification in the U.S. astronauts and two exposure control groups with and without a history of military aviation, respectively. The goals are to determine the effect of space radiation exposure on the severity and progression of age-related cortical (C), nuclear (N), and posterior subcapsular (PSC) lens opacities.

Methods: Astronauts and exposure controls were recruited from the local occupational aerospace community. Severity of C, N, PSC lens opacification, age, demographics and general health, nutritional intake, solar ocular exposure, and other confounding variables were measured at baseline. Results: A total of 419 subjects were enrolled (223 astronauts, 96 military aviator controls, and 100 ground controls) who met inclusion criteria. Significant confounder variables were a history of asthma, history of hypertension, history of obesity, number of drug reported taking, and maximum high contrast LogMAR acuity. Propensity scores based on confounders suggest that military aviator controls were more similar to astronauts when compared with ground controls.

Conclusions: Confounder patterns based on military aviator controls were more similar to astronauts when compared with ground controls. This will be important for dose-response models employed during the cross-sectional and longitudinal analyses.

MANUSCRIPT 2:

TITLE: NASCA Report 2 – Cross-sectional Analyses of Space Radiation Exposure and Risk of Lens Opacity

AUTHORS: Leo T. Chylack, Jr., M.D.(a), Alan H. Feiveson, Ph.D.(b), Leif Peterson, Ph.D.(c), F. Keith Manuel, O.D.(e), Mary L. Wear, Ph.D.(d), William H. Tung, B.S.(a), Dale Hardy, M.S.(c), Lisa Marak, R.N.(d), Cynthia Bell, M.S.(f), Francis Cucinotta, Ph.D.(b)

AFFILIATIONS: a. Center for Ophthalmic Research, Brigham and Women's Hospital, Boston, MA; b) NASA, Johnson Space Center; c) Baylor College of Medicine, Department of Medicine; d) Wyle Laboratories; e) Space Center Eye Associates; f) USRA Division of Space Life Sciences

ABSTRACT: Objectives: NASCA is a five-year investigation in U.S. astronauts to determine the effect of space radiation exposure on severity and progression of cortical (C), nuclear (N), and posterior subcapsular (PSC) lens opacities.

Methods: Objective measures of C, P, and N opacification were obtained from Nidek EAS 1000 digital lens images. Space radiation exposure was expressed as total lens dose/subject aggregated over missions. Solar ocular exposure and nutritional status were characterized. Statistical analyses consisted of fitting customized non-normal regression models to measures of opacity and testing effects of subject group, radiation exposure and other variables on degree of opacity.

Results: Age is the most important predictor of C. Older subjects had higher mean and more varied C. Astronauts who have not yet flown and aircrew controls had less C than controls, while astronauts who flew at least one mission in space had significantly greater C. Increased solar UV was associated with higher, and larger intake of A-vitamins with lower, C. Astronauts had less N, however neither space nor solar radiation exposure was associated with more N. No nutritional factors were associated N. Age was weakly associated with increased P.

Conclusions: Exposure to space radiation increases risk of C opacification.

As of February, 2007 the NASCA project is well into its third year, and when the last of the three-yearm evaluations is completed, we will be able to begin the first set of analyses of the longitudinal phase of the project in which we examine the relationship between the factors of interest to the rate of progression of lens opacification. This will be an important milestone in this project, since these measures are likely to be of greatest interest to NASA and the greater eye and radiation research communities.

POSTER ABSTRACT**:

** PLEASE SEE “RESULTS SECTION” FOR UPDATED FINDINGS WHICH IN SOME CASES DIFFER SIGNIFICANTLY FROM THOSE CITED IN ABSTRACT

Purpose: NASCA is a 5-yr study of the risk factors associated with cataract incidence and progression in US astronauts, military pilots, and ground-based controls (GBC). Risk factors studied include: space radiation exposure, nutrition, UV exposure, and age.

Methods: Areal extent of cortical (C), and posterior subcapsular (P) opacity and density of nuclear (N) opacity were assessed in Nidek EAS 1000 digital lens images. Nuclear color was graded with LOCS III. We used the Harvard Food Frequency Questionnaire (HFFQ). Lens-specific radiation exposures from space, medical sources, aviation training, and isotopes were incorporated. Statistical analyses used space radiation dose-response modeling of baseline prevalence of C, P, and N adjusted for age, nutrition, and solar exposure. Several dose-response models were considered, choosing the most parsimonious ones that described the data. We modeled distribution of log C area opaque as a skew-normal distribution, discrete levels of the area of P opacification with an ordered logit-regression and N density at three loci with beta distributions. In each case we determined the statistical significance of age, space radiation exposure, and 105 nutritional variables on the distribution of each type of opacity.

Results: Complete data from 199 astronauts, 90 pilots, and 95 GBC were available after 1 yr. In our preliminary results baseline prevalences of C, N, and P among astronauts and pilots were significantly lower than in GBCs. For astronauts only, age, radiation and sun exposures were statistically significantly associated with worsening, and a- and ß-carotene with amelioration, of C. For N age was the main explanatory variable for all loci. For astronauts only, space radiation had no effect on pixel density of the central N clear zone or the posterior embryonal nucleus, but it did have a small statistically significant effect (worsening) on the pixel density of the N anterior embryonal nucleus. Lutein had a strong protective effect at each of the three N loci. a- and ß-carotene produced similar results. Age was the only variable statistically significantly associated with extent of P.

Conclusions: Age was significantly associated with severity in C, P, and N. For astronauts only, space radiation and UV exposures were significantly associated with worsening C and with opacification of the anterior embryonal nucleus. a- and ß-carotene and lutein had small beneficial effects on C and N opacification, respectively.

Commercial Relationships: All authors: None.

Support: NASA NAG9-01491.

** PLEASE SEE “RESULTS SECTION” FOR UPDATED FINDINGS WHICH IN SOME CASES DIFFER SIGNIFICANTLY FROM THOSE CITED IN ABSTRACT

NARRATIVE PROGRESS REPORT

PROJECT OBJECTIVES:

1. To determine if exposure to space radiation produces more cortical, posterior subcapsular (PSC), or nuclear opacity in astronauts vs. comparison groups after adjustment for age, solar radiation exposure, and possible nutritional effects. 2. To identify nutritional groups that appear likely to be associated with increased or decreased cortical, PSC, or nuclear opacity.

RECRUITMENT STATISTICS AS OF MARCH 23, 2006:

As of March 23, 3006 we have recruited 222 astronauts (188 male/34 female), 94 military pilots (88 male/6 female), and 99 ground-based controls (79 male/20 female).

METHODS:

We gathered demographic data (age, gender, other) with standardized questionnaires. Ocular data were obtained for all subjects (astronauts and controls) from NASA’s standardized, comprehensive, annual, ocular examination that included best-corrected LogMAR Visual Acuity, and contrast sensitivity function. To estimate space radiation doses to the lens we used space dosimetry, and we made a best estimate of the mission and astronaut-specific lens doses, or lens dose equivalents including the individual contributions from trapped radiation and galactic cosmic rays (GCR). We estimated solar UV exposures from validated demographic data and personal histories. We estimated the color of the lens nucleus with the Lens Opacities Classification System (LOCS III) measure of nuclear color. We gathered nutritional data with the validated Harvard Food Frequency Questionnaire (HFFQ).

We used digital images of the lens obtained with a specialized camera (Nidek EAS 1000 Lens imaging system) to assess the type and severity of lens opacification.

STATISTICAL ANALYSIS:

Our basic statistical objectives were to test if exposure to space radiation produces higher cortical, PSC, or nuclear opacification in astronauts vs. comparison groups after adjustment for age, solar radiation exposure, and possible nutritional effects, and to identify nutritional groups that appear likely to be associated with increased or decreased cortical, PSC or nuclear opacity. In this first year of the study, only one observation per subject was available, so no opacity growth models were considered here. Our dependent variables were maximum image area opaque (OD, OS) one observation per subject for the first year, and our independent (explanatory) variables were: subject group, age, annual solar UV exposure, space radiation exposure, and weekly intake of 105 nutrients. Because the distributions of many of the variables were skewed, we used Skew-normal regression modeling.

1. Analysis of PSC data:

The Maximum PSC opacity (OD, OS) measures were grouped into 5 binned ranges. We used ordered logistic regression model to test for effects of explanatory variables on probability of maximum PSC opacity falling into each of the above bins.

2. Analysis of nuclear cataract data:

We used the log of the maximum pixel density (OD, OS) for three locations (nuclear central clear zone, anterior embryonal and posterior embryonal nuclei) and Beta regression to test for effects of explanatory variables on the mean and variance of the distribution of pixel density at each location.

RESULTS:

Cortical opacity tends to increase with age, and with intake of Omega-3-fatty acids. Cortical opacity tends to be lowered by greater intake of A-vitamins and pro-vitamins. There is increased variation in cortical opacities for older subjects and in those exposed to space radiation. There does not appear to be much evidence supporting more variation in cortical opacities for persons with higher sun exposure. There is considerable skewness in the distribution of log cortical opacities.

Posterior Subcapsular (PSC) Opacity: The only explanatory variable that was significantly associated with PSC opacity was age

Nuclear Cataracts: Increased age was shown to be an overwhelming contributor to density of nuclear cataract at all three measurement loci. No other explanatory variable was found to be significantly associated with pixel density.

CONCLUSIONS:

Cortical Cataracts:

1. Age is the most important predictor of cortical opacification. Populations of older people have a higher mean cortical opacity (P<0.001) and a larger spread of opacities (P = 0.012) than do populations of younger people.

2. Astronauts who have not yet flown and non-astronaut pilots appear to have less mean opacity than the ground-based comparison group (P = 0.10) and more clearly have a tighter distribution of cortical opacities (P = 0.017),

3. The population of astronauts who have been exposed to space radiation (i.e. have flown at least one mission) has a significantly increased variation in opacity relative to the other astronauts and pilots (P = 0.007)

4. Astronauts who have flown do not have a discernable difference in mean opacity from those who have not flown. Within the group of astronauts who have flown, there is no relationship between the total amount of space radiation received and the mean or variance of cortical opacity.

5. Solar radiation exposure had no discernable effect on either the mean or variance of cortical opacity.

6. Nutritional Findings: Larger intake of omega fatty acids appears associated with increased cortical opacity (P = 0.012*). Larger intake of A-vitamins and pro-vitamins appears associated with decreased cortical opacity (P = 0.004*)

PSC and Nuclear Cataracts:

1. Increasing age was significantly associated with increasing opacity. No other significant explanatory variables were found. 2. Although no factors other than age appeared to effect PSC and nuclear cataracts across subjects in this cross-sectional study, there may well be effects of radiation (space and/or solar) on the growth rates of these types of cataracts. This question will be addressed when the longitudinal analysis phase begins after the second year of the study.

MANUSCRIPTS IN PREPARATION:

1. We are preparing a manuscript describing the NASCA methodology and recruitment results after the first year. The paper will also have baseline descriptive data on all of the experimental variables assessed in this project. We expect to complete this manuscript and submit it to Archives of Ophthalmology before July, 2006. 2. We are also preparing a second manuscript describing the results of a cross-sectional analysis of the cataractogenic risks of space radiation for astronauts, pilots, and ground-based controls. We will also present the effect of nutritional variables, age, gender, and annual solar UV exposure on these risks. The results of this analysis are described in the annual report (see earlier sections). We also expect to complete this manuscript and submit it to Archives of Ophthalmology before July, 2006.

3. We are analyzing the relationship between the number of loci in the phenotype of posterior subcapsular cataracts and the dose of space radiation. There are significant findings in this analysis, but we have not yet decided if they will be included in Manuscript 2 (above) or in a separate paper. We will make this decision in the next few weeks.

4. As we move into the longitudinal phase of the NASCA project we expect to be able to assess the relationship between dose of space radiation and measures of cataract incidence (new cases of lens opacification) and progression (of existing opacities) and the effects of age, gender, nutrition, and annual solar UV exposure on these progression rates. The manuscripts for these analyses will probably be cataract-specific (separate papers for cortical, nuclear, and posterior subcapsular cataract).

Respectfully submitted,

Leo T. Chylack, Jr., M.D. Principal Investigator of NASCA Project

Ocular examination and LogMAR acuity measurement: We are grading nuclear color as a means of decreasing the variability of our measurements of nuclear opacification.

Nidek EAS 1000 Lens Imaging Analysis: We have completed the image analyses of 250 subjects who have been through the ocular examination protocol to date. We have encountered no unanticipated difficulties in the image analyses. The images have been of unusually high quality. We expect to receive another 90 subjects-worth of image data on 05/01 and we will have all of these image analyses completed before the end of July 2005 when we will begin the cross-sectional analyses of the NASCA data.

Astronauts: 93 males, 21 females, for a total of 114. Military pilots: 60 males, 2 females, for a total of 62. LSAH subjects: 63 males, 11 females, for a total of 74. Grand Total: 250 (51.2%) We have also made additional progress are as follows:

Human Subjects Committee (HSC) applications: We have completed and secured approval from the HSCs at the Brigham and Women’s Hospital, Baylor College of Medicine, and NASA/JSC.

NASCA Website: At the BWH we have established a secure NASCA-specific server, and we are now uploading administrative information (Manual of Operations, award statements, policies, teleconference minutes, etc.), image data, recruitment statistics, relevant references, and many other study-related materials. The server will also hold a copy of the main NASCA data file prepared by Dr. Wear at Wyle Labs.

Data Management: Dr. Mary Wear and her staff at Wyle labs are compiling the data sent from the various study sites (ocular and general medical examinations, data from analysis of digitized lens images, radiation exposure statistics, nutritional Food Frequency Questionnaires, demographic data, and routine blood data. The team is about to begin the process of assessing the quality of the data gathered to date. The number of variables will be reduced to those with direct primary relevance to the study’s objectives. Lens/Cataract Image Analysis: As of 01/14/05 we have completed the analyses of lens images (slit and retroillumination) from 197 subjects. Also, we have completed the development of protocols for analyzing the different forms of cataract. Statistical Analysis: Dr. Al Feiveson at NASA is currently reviewing the data sets with a goal of developing appropriate statistical models for the initial cross-sectional analyses of the risk factors for cataract prevalence. Radiation Data: A database of radiation doses received by astronauts from all known sources will be used in the NASCA study. The database includes astronaut radiation exposures from medical, aviation training, isotopes, and space radiation. Computer models are used to estimate the lens dose based on available dosimetry for each exposure

Food Frequency Questionnaire Data: At the time of the 2004 Annual Report, the NASCA team had not selected the format for gathering nutritional data. The indecision was due partly to the uncertainty about the feasibility of using blood to assess the nutritional status of astronauts and controls. We were particularly interested in assessing the antioxidant status of individuals in all groups using anti-oxidant indices. Two factors helped us make the decision against using blood-derived data to assess nutritional status: 1) the cost of these analyses would be prohibitively high even if some of them were done “in house” at NASA, and 2) there were respected peer review publications suggesting that blood levels accurately reflect recent nutritional intake but are not robust as indicators of chronic or long-term nutritional patterns. Having made the decision not to use blood analyses to determine anti-oxidant status, we decided to evaluate nutritional questionnaires as alternative methodologies. The Harvard Food Frequency Questionnaire (HFFQ) proved to be a validated instrument that was reasonably efficient to use in the setting of the NASCA study. Accordingly we obtained HSC approval to use the HFFQ and began using this at the outset of the study. Dale Hardy, our Patient Recruiter is administering these questionnaires in a standardized manner and the Channing Laboratory at the Brigham and Women’s Hospital is doing the automated assessment of the nutritional information. Data in electronic form will be sent to Dr. Wear for entry into the NASCA database. Publications: Plans are underway for the first series of publications of the results of the cross-sectional component of the NASCA study. In the spring of 2005 we will have sufficient data to begin these analyses, and we anticipate that we will have the first analytical results in early summer of 2005.