Contents List

• BC - breast cancer (Last updated: 25/08/2000)
• BC data preparation: brief written protocol
• Specification of BC data format (2000 overview)
• BC 2000 data request
• BC 2000 runtime assumptions
• Guide to coding some death causes in EBCTCG 2000 data format
• Breast cancer data directories for 2000 overview (flow chart)
• The procedure for breast cancer data importing and checking (flow chart)
• 2000 Overview Analysis Plan

BC data preparation: brief written protocol

All patient records are converted into the breast cancer data format (2000 overview) (described below), if not already supplied in it. Results received as tables are converted into sets of synthetic records. The following routine checks (where appropriate) are performed on every compilation:

• Duplicate patient entries
• Randomisation date wrong, before 1945 or out of range
• Patient identifier missing
• Second malignancy date wrong or out of range
• Randomisation date missing
• Distant/unknown recurrence date wrong or out of range
• Treatment allocation missing
• Prior local recurrence date wrong or out of range
• Randomisation age missing
• Last follow-up or death date wrong or out of range
• Menopausal status missing
• Treatment allocation code unknown
• Surgery type missing
• Randomisation age not in range 20-89
• Axillary status code missing or unknown
• Menopausal status code unknown
• Estrogen receptor coding missing or unknown
• Surgery code unknown
• Progesterone receptor coding missing or unknown
• Second malignancy site missing
• Estrogen receptor status code missing or unknown
• Second malignancy date missing
• Progesterone receptor status code missing or unknown
• Laterality of primary tumours missing
• Size of primary tumour missing
• Historical grade of primary tumour missing
• Distant/unknown recurrence flag missing
• Second malignancy site code missing or unknown
• ICD revision of second malignancy missing
• Distant/unknown recurrence date missing
• Distant/unknown recurrence flag error
• Prior local recurrence flag missing
• Prior local recurrence flag error
• Prior local recurrence date missing
• Survival status code unknown or missing
• ICD revision for cause of death missing or unknown
• Death date missing
• Code for cause of death missing or unknown
• ICD value of death missing
• Cause of death given when alive

If an event such as the recurrence of disease is reported at a date later than the quoted last follow-up date, the last follow-up date is automatically changed to the later date. The completeness of follow-up is then calculated for the end of each calendar year. The distributions of randomisation dates, randomisation ages and time elapsed since last follow-up are checked for any significant imbalance between treatment groups in two ways as follows. Firstly, a t-test is applied to the difference between the mean value of each distribution for patients in each group with the corresponding mean for patients in the remainder. Secondly, an F-ratio is calculated for each distribution by comparing the variance between the groups with the variance within the groups. The distribution of time elapsed since last follow-up is also checked in these two ways for any significant imbalance between those patients with and those patients without a recorded recurrence of disease. Finally, the distribution of time elapsed since last follow-up is checked in the same two ways for any significant imbalance between patients in two categories of menopausal status (pre- or perimenopausal; postmenopausal), two categories of axillary nodal status (negative; positive), two categories of oestrogen receptor status (poor; positive) and two categories of progesterone receptor status (poor; positive).

Where patient serial numbers form an obvious sequence it is checked for missing numbers.

A tabulated breakdown of variables is produced for each trial, together (where relevant) with lists of patents in 'problematical' categories such as those with lapsed follow-up, uncertain death cause or second malignancy site. Graphs of accrual date and the proportion of living patients still on follow-up as a function of time from randomisation by treatment allocation are also produced, together with Kaplan-Meier life-table curves. Before trial data are finally incorporated into the overview, the analyses described above are sent to the participating trialist(s) for checking and approval.

Contact

Vaughan Evans
EBCTCG Secretariat
C.T.S.U.
Radcliffe Infirmary
Oxford OX2 6HE
England
Tel. U.K.(44)-Oxford(1865)-404852
FAX: U.K.(44)-Oxford(1865)-404853
Email: vaughan.evans@ctsu.ox.ac.uk

Specification of BC data format (2000 overview)

BC 2000 Data Request

EBCTCG Year 2000 data format

BASELINE VARIABLES

Patient identifier (up to 12 characters)

Date of randomisation (please specify your format for dates)

Allocated treatment (please specify your codes)

Age at randomisation (years)

Menopausal status at randomisation (1=pre-menopausal, 2=peri-menopausal, 3=post-menopausal)

Original surgery (Please use and specify your codes; or use 1=radical mastectomy, 2=total mastectomy, 3=simple mastectomy, 4=partial mastectomy with axillary clearance, 5=partial mastectomy without axillary clearance, 6=lumpectomy with axillary clearance, 7=lumpectomy without axillary clearance)

Axillary status at randomisation (Please use and specify your codes; or use 1=N0 clearance, 2=N1-3 clearance, 3=N4+ clearance, 4=N- sample, 5=N+ sample, 6=N- clinical, 7=N+ clinical)

Estrogen receptor (ER) coding (Please use the following codes to indicate the units used for the ER measurement: F=fmol/mg cytosol protein, P=percent cells staining, C=other codes used for ER protein result [please specify codes]). Please leave blank if there is no ER value for this woman.

ER status of primary tumour at the time of original surgery

Progesterone receptor (PR) coding (Please use the following codes to indicate the units used for the PR measurement: F=fmol/mg cytosol protein, P=percent cells staining, C=other codes used for PR protein result [please specify codes]). Please leave blank if there is no PR value for this woman.

PR status of primary tumour at the time of original surgery

Laterality of primary tumour(s) at the time of randomisation (1=left, 2=right, 3=bilateral)

Largest recorded diameter of primary tumour at the time of original surgery (Please use millimetres [so, for example, 2cm is 20] or use and specify your codes)

Histological grade of primary tumour (Please use and specify your codes; or use 1=well differentiated, 2=moderately differentiated, 3=poorly differentiated)

FOLLOW-UP VARIABLES

Site of second malignancy including second breast cancers (Please use and specify your codes)
(If you use the WHO's international Classification of Disease [ICD] codes, please specify which ICD revision [e.g. 8 or 9] is used)

Date of diagnosis of second malignancy (please specify your format for dates)

Any distant recurrence or recurrence of unknown site? (1=no, 2=distant, 3=unknown site)

Date of this recurrence (please specify your format for dates)

Any isolated local recurrence prior to this recurrence? (Please try to distinguish ipsilateral breast recurrences from other local recurrences. Use and specify you own codes; or use 1=no local recurrence, 2=any local recurrence, 3=ipsilateral breast recurrence,  4=other locoregional site)

Date of prior local recurrence (please specify your format for dates)

Status (1=alive, 2=died)

Date last seen or died (please specify your format for dates)

Cause of death for patients dying without distant recurrence (Please use and specify your codes)        (If you use ICD codes, refer to the notes to cols 56-62)

NOTES ON FORMAT OF DATA REQUEST

Dates that are not (or not yet) known exactly

either leave DAY blank, and give (approximate or provisional) month and year;
or leave DAY and MONTH blank, and just give approximate year.

Date randomised

Data are requested for ALL patients EVER randomised, including even lost or withdrawn patients, and ignore all non-randomised patients.

Treatment group allocated

Treatment group number: 1 or 2 only, for 2-group trials, or a wider range for trials with more arms, as defined by trialist. N.B: even if, in reality, some quite different (or even opposite!) treatment was inadvertently given, what is wanted is the originally-allocated treatment. (For patients erroneously entered more than once, give only the first allocation is requested.)

Menopausal status on randomisation

3 = post-menopausal (includes bilateral ovarian ablation before randomisation). Use trialists' own definitions of menopausal status at entry.

Original surgery

2 = total mastectomy (i.e. with axillary clearance); 3 = simple mastectomy (i.e. without axillary clearance).

BC 2000 Runtime Assumptions

Guide to coding some death causes in EBCTCG 2000 data format

Notes for processing data supplied by trialists

1. If a textual description of the cause of death is given, this should be put in the comments field using the (d. …..) format. If the description includes labels (1a, 1b, etc), these should also be entered as they help to determine the underlying cause of death.

2. If a woman does not have a ‘recurrence’ recorded before her death (ie she has had neither an isolated local recurrence, nor a distant recurrence or recurrence of unknown site, nor a contralateral breast cancer), the CTSU-COD needs to be put into the data. This code should be set to the underlying cause of death using the standard ICD rules for selecting the underlying cause, except that pulmonary embolism should take precedence over the cause of the pulmonary embolism. For deaths after a ‘recurrence’, the CTSU-COD does not need to be put into the data.

3. If the trialist supplies a single ICD code for the underlying cause of death, D.ICDRev and ICD-COD should be put into the data. If a 3 digit code has been supplied, this should be entered as three digits and left justified. If the trialist has supplied more than one ICD code for a woman’s death, and the underlying cause of death is not clear, all the ICD codes should be put in the comments field using the ICDx[xxx,xxx,…] format. If the trialist has not supplied an ICD code, do not spend time trying to assign one.

4. 199.9 (ICD revisions 7, 8, and 9) and C80.9 (ICD revision 10) are special CTSU-invented "ICD codes" to indicate that the trialist hasd told us that they do not have information on the site of malignancy.

5. External causes of death should be coded as being due to the external cause rather than the nature of the injury (i.e. E codes in revisions 7, 8, and 9 and V, W, X and Y codes in ICD 10) where possible.

6. CTSU-CODs 10 and 24 are only used if no better information is available to the trialist.

7. CTSU-COD 1 (iatrogenic) is no longer used. These deaths are now coded to their appropriate place according to ICD rules (eg in ICD9 death due to toxicity from antineoplastic drugs is E933.1, death due to acute myeloid leukaemia probably caused by antineoplastic drugs is 205.0, death due to pulmonary fibrosis that is possibly radiation-induced is 515)

EBCTCG 4TH CYCLE 2000 ANALYSIS PLAN

KEY QUESTIONS

Hormonal

Chemotherapy

Local Therapy

No Analyses Planned

For more detailed information on the analyses see the MS Word document:
OVIEWS ON `SERV1\PUBLIC\STUDIES'\BREAST\4THCYCLE 2000\EBCTCG 4TH CYCLE 2000 ANALYSIS PLAN.DOC.

Last updated 25th May 2000 by Paul McGale