StatCrunch logo (home)

Report Properties
Thumbnail:

from Flickr
Created: Sep 13, 2019
Share: yes
Views: 39
Tags:
 
Results in this report
 
Data sets in this report
None
 
Need help?
To copy selected text, right click to Copy or choose the Copy option under your browser's Edit menu. Text copied in this manner can be pasted directly into most documents with formatting maintained.
To copy selected graphs, right click on the graph to Copy. When pasting into a document, make sure to paste the graph content rather than a link to the graph. For example, to paste in MS Word choose Edit > Paste Special, and select the Device Independent Bitmap option.
You can now also Mail results and reports. The email may contain a simple link to the StatCrunch site or the complete output with data and graphics attached. In addition to being a great way to deliver output to someone else, this is also a great way to save your own hard copy. To try it out, simply click on the Mail link.
LSA Measurements in Women
Mail   Print   Twitter   Facebook

Introduction

Lipid-bound sialic acid (LSA) is a blood maker that can help a clinician and a patient gain invaluable knowledge in assessing overall states of health, specifically related to the presence or absence of cancer cells. LSA elevations in blood have previously been reported in patients with mammary, gastroenteric, pulmonary and ovarian cancers as well as leukemia, lymphoma, melanoma, sarcoma, and Hodgkin disease. (LabCorp, 2019) While LSA does not have the specificity or sensitivity necessary for cancer detection it has been linked in being useful in monitoring the course of therapy and detecting disease reoccurrence in certain cancer patients. In this study, we wish to asses the nature of differences in average LSA measurements among four disease classifications.

Data/Methods

Samples of 100 women for each of the four disease classifications were taken; Group A – normal controls; Group B – patients with benign breast disease; Group C – patients with primary breast cancer; Group D – patients with recurrent metastatic breast cancer. For each woman’s sample, a serum lipid-bound sialic acid measurement was taken in mg/dL.

Analysis

First, we want to check normality and constant variability assumptions. Normality is assumed because the points shown do not deviate too much from the line in the QQ plot as well as equal variability is assumed because the horizontal spread of points is similar. Independence is also assumed due to our patients having no connections to other patients in the sample.

Result 1: QQ Plot of Residuals   [Info]
Right click to copy

Result 2: Group Means vs. Residuals   [Info]
Right click to copy

A one factor ANOVA was then used to determine if there were differences in average LSA measurements among the four disease classifications. The results of the one factor ANOVA confirm that there are differences in averages of LSA measurements in at least two of the disease classifications. (F = 96.4, p-value <0.0001). With these results, we may proceed to a multiple comparison Tukey test to assess the sizes of the differences.

Result 3: One Way ANOVA   [Info]

Analysis of Variance results:


Responses: LSA
Factors: Group

Response statistics by factor


GroupnMeanStd. Dev.Std. Error
A10016.2492.6840530.2684053
B10020.482.9479320.2947932
C10019.9763.01159780.30115978
D10023.2923.13083980.31308398

ANOVA table


SourceDFSSMSF-StatP-value
Group32513.8239837.9412996.40839<0.0001
Error3963441.86598.6915806
Total3995955.6898

Tukey HSD results (95% level)


A subtracted from
DifferenceLowerUpperP-value
B4.2313.15532895.3066711<0.0001
C3.7272.65132894.8026711<0.0001
D7.0435.96732898.1186711<0.0001
B subtracted from
DifferenceLowerUpperP-value
C-0.504-1.57967110.571671050.6216
D2.8121.73632893.8876711<0.0001
C subtracted from
DifferenceLowerUpperP-value
D3.3162.24032894.3916711<0.0001

With 95% confidence we estimate that the average LSA measurements are greater for women in Group B, C and D than they are for women in Group A, anywhere from 2.65mg/dL to 8.19mg/dL higher.

With 95% confidence we estimate that the average LSA measurements are 1.74mg/dL to 3.89mg/dL greater for women in Group D than for women in Group B.

With 95% confidence we estimate that the average LSA measurements are 2.24mg/dL to 4.39mg/dL greater in women in Group D than for women in Group C.

Due to 0 being in the confidence interval between Group B and Group C, there is no evidence of a difference in LSA measurements between those two groups.

Conclusion

The results from our study support our suspicions that there are differences in average LSA measurements among four disease classification groups. We found that Group A likely has the lowest average LSA measurements while Group D likely has the highest average LSA measurements. Ultimately, we may conclude that lower LSA levels, on average, are associated with non-cancerous diseases while higher levels, on average, are associated with reoccurring metastatic breast cancer. There may be some difference between Group B and Group C, however our study was not appropriately powered to detect this difference if in fact it does exist. This is concerning clinically because Group B represents women with benign breast disease while Group C represents women with primary breast cancer. Future studies could aim to have larger sample sizes for these two groups, benign breast disease and primary breast cancer, to better determine the difference in LSA measurements between these groups while monitoring therapy or detecting disease reoccurrence.

References

LabCorp. (2019). Lipid-associated Sialic Acid. Retrieved from LabCorp: https://www.labcorp.com/test-menu/30416/lipid-associated-sialic-acid-lasa

 

 

HTML link:
<A href="https://www.statcrunch.com/5.0/viewreport.php?reportid=90110">LSA Measurements in Women</A>

Comments
Want to comment? Subscribe
Already a member? Sign in.
By nku.dr.nolan
Sep 15, 2019

Hi Kelsey - a few comments:

1. In Data/Methods, don't forget to talk about what methods will be used. Analysis then applies those methods.
2. You did a good job with your statistical interpretations. Maybe initially do not combine the CI's for A vs B, A vs C, and A vs D. But in the end/discussion I think you do want to make overall statements.
3. B vs C, any difference in averages that might exist is estimated to be no bigger than 1.57. Only call for "not enough power" if that value would be relevant.

Hope this helps!
By brian.hargis1337
Sep 13, 2019

Your QQ-plot was much more "tight" along the line than I noticed in my report, which had 113 participants, and the top end of the graph appeared to have more outliers than with your data points in this report. It seems like the largest group does have a large residual when compared to its smallest group, it seemed like my report had less extremes, but that could be the nature of the LSA measurements being a more "pass/fail" accruement within the body to be detected in the first place, in the presence of cancer.
By laura.boettcher24
Sep 13, 2019

Kelsey,
Great report! I must admit I have never heard of LSA. I wonder if it is a newer test that some laboratories haven't added yet. It looks as if your study showed on average that higher LSA levels were associated with reoccurring metastatic breast cancer. It would be interesting to see what a larger sample size would do to the results. Also, I think it should look at other types of cancers too. This could be a great test for physicians to utilize for their patients.

Always Learning