Thrombocytosis: an underused risk marker of cancer in ...



Thrombocytosis: an underused risk marker of cancer in primary care?Lay summaryBackground: The UK does very badly in cancer, and any way of diagnosing it more quickly would be welcomed.The UK has a poor record in cancer outcomes, much of which is blamed on delays in diagnosis. Many initiatives have been established to help primary care identify and investigate patients with possible cancer. One factor which has received little attention is a raised platelet count (thrombocytosis). Our team has identified this as being relevant in lung, colon, ovary, oesophago-gastric, bladder and pancreas cancers. Together, these six cancer sites suggest that the risk of an underlying cancer in a patient with thrombocytosis exceeds 4% and may be as high as 10%. Such a figure would make a very important contribution to cancer diagnosis: a contribution that is currently completely absent from current diagnostic practice. Research objective: How useful is thrombocytosis in predicting an underlying cancer?In all our previous studies, we have started with a cancer and worked backwards towards clinical features including thrombocytosis. In this study, we will start with general practice patients with thrombocytosis, and identify a) which cancers are associated, b) how strong the association is – in terms of a percentage risk, and c) which level of platelet count is optimal for recommending cancer investigation (incorporating the patient’s age and sex). The study will use 50,000 patients from the CPRD, 40,000 with a raised platelet count, and 10,000 without. We will link this dataset to the national cancer registries to confirm cancer diagnoses. We will perform a cohort study, with sub-analyses by age, sex and platelet count. Our power calculation has shown that 50,000 patients will provide sufficient cases to give very reliable results.Main outcome measure: the incidence of cancer in the one year following a raised platelet count. Our team is extremely experienced in both CPRD and cancer diagnostic studies, containing a statistician, professors of diagnostic medicine and a practising haematologist. Furthermore, the research group into which this study will be embedded has an extremely strong record in clinical implementation of research findings. BackgroundEarly diagnosis of cancer is a leading objective, particularly in those countries like the UK and Denmark with relatively poor cancer outcomes. ADDIN EN.CITE <EndNote><Cite><Author>Coleman</Author><Year>2011</Year><RecNum>2073</RecNum><DisplayText>(1)</DisplayText><record><rec-number>2073</rec-number><foreign-keys><key app="EN" db-id="pvparxz00xww0res2pdxras85e952wf50xwf">2073</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Coleman, M. P.</author><author>Forman, D.</author><author>Bryant, H.</author><author>Butler, J.</author><author>Rachet, B.</author><author>Maringe, C.</author><author>Nur, U.</author><author>Tracey, E.</author><author>Coory, M.</author><author>Hatcher, J.</author><author>McGahan, C. E.</author><author>Turner, D.</author><author>Marrett, L.</author><author>Gjerstorff, M. L.</author><author>Johannesen, T. B.</author><author>Adolfsson, J.</author><author>Lambe, M.</author><author>Lawrence, G.</author><author>Meechan, D.</author><author>Morris, E. J.</author><author>Middleton, R.</author><author>Steward, J.</author><author>Richards, M. A.</author></authors></contributors><titles><title>Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995-2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data</title><secondary-title>The Lancet</secondary-title></titles><periodical><full-title>The Lancet</full-title></periodical><pages>127-138</pages><volume>377</volume><number>9760</number><keywords><keyword>general Cancer in PC</keyword></keywords><dates><year>2011</year></dates><isbn>0140-6736</isbn><urls><related-urls><url>: 10.1016/s0140-6736(10)62231-3</electronic-resource-num><access-date>2011/1/14/</access-date></record></Cite></EndNote>(1) Screening helps, although it is only effective in a small number of cancers. Because of this, most cancers (90% in the UK) present with symptoms – and usually to primary care. ADDIN EN.CITE <EndNote><Cite><Author>Hamilton</Author><Year>2010</Year><RecNum>2062</RecNum><DisplayText>(2)</DisplayText><record><rec-number>2062</rec-number><foreign-keys><key app="EN" db-id="pvparxz00xww0res2pdxras85e952wf50xwf">2062</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Hamilton, William</author></authors></contributors><auth-address>NIHR School for Primary Care Research, Department of Community Based Medicine, University of Bristol, Bristol. w.hamilton@bristol.ac.uk</auth-address><titles><title>Cancer diagnosis in primary care</title><secondary-title>British Journal of General Practice</secondary-title></titles><periodical><full-title>British Journal of General Practice</full-title></periodical><pages>121-8</pages><volume>60</volume><number>571</number><keywords><keyword>general Cancer in PC</keyword></keywords><dates><year>2010</year><pub-dates><date>Feb</date></pub-dates></dates><accession-num>20132704</accession-num><urls><related-urls><url> Technologies</remote-database-provider></record></Cite></EndNote>(2) Many initiatives have been implemented to expedite diagnosis, but most have targeted better investigative services, after the general practitioner has considered the possibility of cancer. These include the two-week wait clinics, open access GP investigation and NICE guidance. Much less attention has focussed on an earlier stage in primary care: triggering the GP to consider cancer at all. Although some cancers have classical symptoms, such as a breast lump in breast cancer and haematuria in urological cancers, many cancers present with vaguer symptoms, often called the ‘low-risk-but-not-no-risk symptoms’. With these presentations, the GP may not consider cancer at all, or may decide against investigation – or be discouraged from investigation by NICE guidance. ADDIN EN.CITE <EndNote><Cite><Author>NICE</Author><Year>2005</Year><RecNum>1072</RecNum><DisplayText>(3)</DisplayText><record><rec-number>1072</rec-number><foreign-keys><key app="EN" db-id="pvparxz00xww0res2pdxras85e952wf50xwf">1072</key></foreign-keys><ref-type name="Book">6</ref-type><contributors><authors><author>NICE</author></authors></contributors><titles><title>Referral Guidelines for suspected cancer</title></titles><keywords><keyword>General cancer in PC</keyword></keywords><dates><year>2005</year><pub-dates><date>June 2005</date></pub-dates></dates><pub-location>London</pub-location><publisher>NICE</publisher><urls><related-urls><url>;(3) This situation can be improved by tables of risk profiles (all English GPs have been sent risk profiles for the common symptom clusters in lung and colorectal cancer) or by computerised support. Such support can work in two ways: either by searching for additional symptoms of cancer once a symptom has been entered into the GP’s clinical system, and calculating an overall risk, or by regular sweeps of the practice computer for symptom profiles deemed to represent a high risk. ADDIN EN.CITE <EndNote><Cite><Author>Hamilton</Author><Year>2012</Year><RecNum>2181</RecNum><DisplayText>(4)</DisplayText><record><rec-number>2181</rec-number><foreign-keys><key app="EN" db-id="pvparxz00xww0res2pdxras85e952wf50xwf">2181</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Hamilton, W</author></authors></contributors><titles><title>Computer assisted diagnosis of ovarian cancer in primary care</title><secondary-title>BMJ</secondary-title></titles><periodical><full-title>BMJ</full-title></periodical><pages>10.1136/bmj.d7628</pages><volume>344</volume><keywords><keyword>Ovarian cancer</keyword></keywords><dates><year>2012</year></dates><urls></urls></record></Cite></EndNote>(4) These schemes are still in their infancy. Platelets and cancerOne overlooked marker of possible malignancy is a raised platelet count, or thrombocytosis. The association between thrombocytosis and cancer was first reported over a century ago. 3.1 Secondary care studies. An old study from a US hospital reported that 30 of 82 patients with a platelet count over 400,000 per mm3 had a malignancy, including stomach, colon, lung, breast and ovarian cancers. ADDIN EN.CITE <EndNote><Cite><Author>Levin</Author><Year>1964</Year><RecNum>2182</RecNum><DisplayText>(5)</DisplayText><record><rec-number>2182</rec-number><foreign-keys><key app="EN" db-id="pvparxz00xww0res2pdxras85e952wf50xwf">2182</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Levin, J</author><author>Conley, CL</author></authors></contributors><titles><title>Thrombocytosis associated with malignant disease</title><secondary-title>Archives of Internal Medicine</secondary-title></titles><periodical><full-title>Archives of Internal Medicine</full-title></periodical><pages>497-500</pages><volume>114</volume><keywords><keyword>general Cancer in PC</keyword></keywords><dates><year>1964</year></dates><urls></urls></record></Cite></EndNote>(5) Another hospital study reported thrombocytosis in lung, but not colon cancer. ADDIN EN.CITE <EndNote><Cite><Author>Costantini</Author><Year>1990</Year><RecNum>1106</RecNum><DisplayText>(6)</DisplayText><record><rec-number>1106</rec-number><foreign-keys><key app="EN" db-id="pvparxz00xww0res2pdxras85e952wf50xwf">1106</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Costantini, V.</author><author>Zacharski, L. R.</author><author>Moritz, T. E.</author><author>Edwards, R. L.</author></authors></contributors><titles><title>The platelet count in carcinoma of the lung and colon</title><secondary-title>Thrombosis &amp; Haemostasis.</secondary-title></titles><pages>501-5</pages><volume>64</volume><number>4</number><keywords><keyword>Lung</keyword></keywords><dates><year>1990</year></dates><urls></urls></record></Cite></EndNote>(6) Recent laboratory studies of the biology of thrombopoiesis in ovarian cancer have demonstrated interleukin-6 production by the tumour leading to increased thrombopoetin synthesis in the liver.PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5TdG9uZTwvQXV0aG9yPjxZZWFyPjIwMTI8L1llYXI+PFJl

Y051bT4yMTgzPC9SZWNOdW0+PERpc3BsYXlUZXh0Pig3KTwvRGlzcGxheVRleHQ+PHJlY29yZD48

cmVjLW51bWJlcj4yMTgzPC9yZWMtbnVtYmVyPjxmb3JlaWduLWtleXM+PGtleSBhcHA9IkVOIiBk

Yi1pZD0icHZwYXJ4ejAweHd3MHJlczJwZHhyYXM4NWU5NTJ3ZjUweHdmIj4yMTgzPC9rZXk+PC9m

b3JlaWduLWtleXM+PHJlZi10eXBlIG5hbWU9IkpvdXJuYWwgQXJ0aWNsZSI+MTc8L3JlZi10eXBl

Pjxjb250cmlidXRvcnM+PGF1dGhvcnM+PGF1dGhvcj5TdG9uZSwgUmViZWNjYSBMLjwvYXV0aG9y

PjxhdXRob3I+TmljaywgQWxwYSBNLjwvYXV0aG9yPjxhdXRob3I+TWNOZWlzaCwgSWFpbiBBLjwv

YXV0aG9yPjxhdXRob3I+QmFsa3dpbGwsIEZyYW5jZXM8L2F1dGhvcj48YXV0aG9yPkhhbiwgSGVl

IERvbmc8L2F1dGhvcj48YXV0aG9yPkJvdHRzZm9yZC1NaWxsZXIsIEp1c3RpbjwvYXV0aG9yPjxh

dXRob3I+UnVwYWltb29sZSwgUmFqZXNoYTwvYXV0aG9yPjxhdXRob3I+QXJtYWl6LVBlbmEsIEd1

aWxsZXJtbyBOLjwvYXV0aG9yPjxhdXRob3I+UGVjb3QsIENoYWQgVi48L2F1dGhvcj48YXV0aG9y

PkNvd2FyZCwgSmVybWFpbmU8L2F1dGhvcj48YXV0aG9yPkRlYXZlcnMsIE1pY2hhZWwgVC48L2F1

dGhvcj48YXV0aG9yPlZhc3F1ZXosIEhlcm5hbiBHLjwvYXV0aG9yPjxhdXRob3I+VXJiYXVlciwg

RGlhbmE8L2F1dGhvcj48YXV0aG9yPkxhbmRlbiwgQ2hhcmxlcyBOLjwvYXV0aG9yPjxhdXRob3I+

SHUsIFdlaTwvYXV0aG9yPjxhdXRob3I+R2Vyc2hlbnNvbiwgSGFubmFoPC9hdXRob3I+PGF1dGhv

cj5NYXRzdW8sIEtvamk8L2F1dGhvcj48YXV0aG9yPlNoYWh6YWQsIE1pYW4gTS4gSy48L2F1dGhv

cj48YXV0aG9yPktpbmcsIEVyaW4gUi48L2F1dGhvcj48YXV0aG9yPlRla2VkZXJlbGksIElicmFo

aW08L2F1dGhvcj48YXV0aG9yPk96cG9sYXQsIEJ1bGVudDwvYXV0aG9yPjxhdXRob3I+QWhuLCBF

ZHdhcmQgSC48L2F1dGhvcj48YXV0aG9yPkJvbmQsIFZpcmdpbmlhIEsuPC9hdXRob3I+PGF1dGhv

cj5XYW5nLCBSdWk8L2F1dGhvcj48YXV0aG9yPkRyZXcsIEFuZ2VsYSBGLjwvYXV0aG9yPjxhdXRo

b3I+R3VzaGlrZW4sIEZyYW5jaXNjYTwvYXV0aG9yPjxhdXRob3I+Q29sbGlucywgS2F0aGVyaW5l

PC9hdXRob3I+PGF1dGhvcj5EZUdlZXN0LCBLb2VuPC9hdXRob3I+PGF1dGhvcj5MdXRnZW5kb3Jm

LCBTdXNhbiBLLjwvYXV0aG9yPjxhdXRob3I+Q2hpdSwgV2FoPC9hdXRob3I+PGF1dGhvcj5Mb3Bl

ei1CZXJlc3RlaW4sIEdhYnJpZWw8L2F1dGhvcj48YXV0aG9yPkFmc2hhci1LaGFyZ2hhbiwgVmFo

aWQ8L2F1dGhvcj48YXV0aG9yPlNvb2QsIEFuaWwgSy48L2F1dGhvcj48L2F1dGhvcnM+PC9jb250

cmlidXRvcnM+PHRpdGxlcz48dGl0bGU+UGFyYW5lb3BsYXN0aWMgVGhyb21ib2N5dG9zaXMgaW4g

T3ZhcmlhbiBDYW5jZXI8L3RpdGxlPjxzZWNvbmRhcnktdGl0bGU+TmV3IEVuZ2xhbmQgSm91cm5h

bCBvZiBNZWRpY2luZTwvc2Vjb25kYXJ5LXRpdGxlPjwvdGl0bGVzPjxwZXJpb2RpY2FsPjxmdWxs

LXRpdGxlPk5ldyBFbmdsYW5kIEpvdXJuYWwgb2YgTWVkaWNpbmU8L2Z1bGwtdGl0bGU+PC9wZXJp

b2RpY2FsPjxwYWdlcz42MTAtNjE4PC9wYWdlcz48dm9sdW1lPjM2Njwvdm9sdW1lPjxudW1iZXI+

NzwvbnVtYmVyPjxrZXl3b3Jkcz48a2V5d29yZD5PdmFyaWFuIGNhbmNlcjwva2V5d29yZD48L2tl

eXdvcmRzPjxkYXRlcz48eWVhcj4yMDEyPC95ZWFyPjxwdWItZGF0ZXM+PGRhdGU+MjAxMi8wMi8x

NjwvZGF0ZT48L3B1Yi1kYXRlcz48L2RhdGVzPjxwdWJsaXNoZXI+TWFzc2FjaHVzZXR0cyBNZWRp

Y2FsIFNvY2lldHk8L3B1Ymxpc2hlcj48aXNibj4wMDI4LTQ3OTM8L2lzYm4+PHVybHM+PHJlbGF0

ZWQtdXJscz48dXJsPmh0dHA6Ly9keC5kb2kub3JnLzEwLjEwNTYvTkVKTW9hMTExMDM1MjwvdXJs

PjwvcmVsYXRlZC11cmxzPjwvdXJscz48ZWxlY3Ryb25pYy1yZXNvdXJjZS1udW0+MTAuMTA1Ni9O

RUpNb2ExMTEwMzUyPC9lbGVjdHJvbmljLXJlc291cmNlLW51bT48YWNjZXNzLWRhdGU+MjAxMi8w

Mi8yMjwvYWNjZXNzLWRhdGU+PC9yZWNvcmQ+PC9DaXRlPjwvRW5kTm90ZT4A

ADDIN EN.CITE PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5TdG9uZTwvQXV0aG9yPjxZZWFyPjIwMTI8L1llYXI+PFJl

Y051bT4yMTgzPC9SZWNOdW0+PERpc3BsYXlUZXh0Pig3KTwvRGlzcGxheVRleHQ+PHJlY29yZD48

cmVjLW51bWJlcj4yMTgzPC9yZWMtbnVtYmVyPjxmb3JlaWduLWtleXM+PGtleSBhcHA9IkVOIiBk

Yi1pZD0icHZwYXJ4ejAweHd3MHJlczJwZHhyYXM4NWU5NTJ3ZjUweHdmIj4yMTgzPC9rZXk+PC9m

b3JlaWduLWtleXM+PHJlZi10eXBlIG5hbWU9IkpvdXJuYWwgQXJ0aWNsZSI+MTc8L3JlZi10eXBl

Pjxjb250cmlidXRvcnM+PGF1dGhvcnM+PGF1dGhvcj5TdG9uZSwgUmViZWNjYSBMLjwvYXV0aG9y

PjxhdXRob3I+TmljaywgQWxwYSBNLjwvYXV0aG9yPjxhdXRob3I+TWNOZWlzaCwgSWFpbiBBLjwv

YXV0aG9yPjxhdXRob3I+QmFsa3dpbGwsIEZyYW5jZXM8L2F1dGhvcj48YXV0aG9yPkhhbiwgSGVl

IERvbmc8L2F1dGhvcj48YXV0aG9yPkJvdHRzZm9yZC1NaWxsZXIsIEp1c3RpbjwvYXV0aG9yPjxh

dXRob3I+UnVwYWltb29sZSwgUmFqZXNoYTwvYXV0aG9yPjxhdXRob3I+QXJtYWl6LVBlbmEsIEd1

aWxsZXJtbyBOLjwvYXV0aG9yPjxhdXRob3I+UGVjb3QsIENoYWQgVi48L2F1dGhvcj48YXV0aG9y

PkNvd2FyZCwgSmVybWFpbmU8L2F1dGhvcj48YXV0aG9yPkRlYXZlcnMsIE1pY2hhZWwgVC48L2F1

dGhvcj48YXV0aG9yPlZhc3F1ZXosIEhlcm5hbiBHLjwvYXV0aG9yPjxhdXRob3I+VXJiYXVlciwg

RGlhbmE8L2F1dGhvcj48YXV0aG9yPkxhbmRlbiwgQ2hhcmxlcyBOLjwvYXV0aG9yPjxhdXRob3I+

SHUsIFdlaTwvYXV0aG9yPjxhdXRob3I+R2Vyc2hlbnNvbiwgSGFubmFoPC9hdXRob3I+PGF1dGhv

cj5NYXRzdW8sIEtvamk8L2F1dGhvcj48YXV0aG9yPlNoYWh6YWQsIE1pYW4gTS4gSy48L2F1dGhv

cj48YXV0aG9yPktpbmcsIEVyaW4gUi48L2F1dGhvcj48YXV0aG9yPlRla2VkZXJlbGksIElicmFo

aW08L2F1dGhvcj48YXV0aG9yPk96cG9sYXQsIEJ1bGVudDwvYXV0aG9yPjxhdXRob3I+QWhuLCBF

ZHdhcmQgSC48L2F1dGhvcj48YXV0aG9yPkJvbmQsIFZpcmdpbmlhIEsuPC9hdXRob3I+PGF1dGhv

cj5XYW5nLCBSdWk8L2F1dGhvcj48YXV0aG9yPkRyZXcsIEFuZ2VsYSBGLjwvYXV0aG9yPjxhdXRo

b3I+R3VzaGlrZW4sIEZyYW5jaXNjYTwvYXV0aG9yPjxhdXRob3I+Q29sbGlucywgS2F0aGVyaW5l

PC9hdXRob3I+PGF1dGhvcj5EZUdlZXN0LCBLb2VuPC9hdXRob3I+PGF1dGhvcj5MdXRnZW5kb3Jm

LCBTdXNhbiBLLjwvYXV0aG9yPjxhdXRob3I+Q2hpdSwgV2FoPC9hdXRob3I+PGF1dGhvcj5Mb3Bl

ei1CZXJlc3RlaW4sIEdhYnJpZWw8L2F1dGhvcj48YXV0aG9yPkFmc2hhci1LaGFyZ2hhbiwgVmFo

aWQ8L2F1dGhvcj48YXV0aG9yPlNvb2QsIEFuaWwgSy48L2F1dGhvcj48L2F1dGhvcnM+PC9jb250

cmlidXRvcnM+PHRpdGxlcz48dGl0bGU+UGFyYW5lb3BsYXN0aWMgVGhyb21ib2N5dG9zaXMgaW4g

T3ZhcmlhbiBDYW5jZXI8L3RpdGxlPjxzZWNvbmRhcnktdGl0bGU+TmV3IEVuZ2xhbmQgSm91cm5h

bCBvZiBNZWRpY2luZTwvc2Vjb25kYXJ5LXRpdGxlPjwvdGl0bGVzPjxwZXJpb2RpY2FsPjxmdWxs

LXRpdGxlPk5ldyBFbmdsYW5kIEpvdXJuYWwgb2YgTWVkaWNpbmU8L2Z1bGwtdGl0bGU+PC9wZXJp

b2RpY2FsPjxwYWdlcz42MTAtNjE4PC9wYWdlcz48dm9sdW1lPjM2Njwvdm9sdW1lPjxudW1iZXI+

NzwvbnVtYmVyPjxrZXl3b3Jkcz48a2V5d29yZD5PdmFyaWFuIGNhbmNlcjwva2V5d29yZD48L2tl

eXdvcmRzPjxkYXRlcz48eWVhcj4yMDEyPC95ZWFyPjxwdWItZGF0ZXM+PGRhdGU+MjAxMi8wMi8x

NjwvZGF0ZT48L3B1Yi1kYXRlcz48L2RhdGVzPjxwdWJsaXNoZXI+TWFzc2FjaHVzZXR0cyBNZWRp

Y2FsIFNvY2lldHk8L3B1Ymxpc2hlcj48aXNibj4wMDI4LTQ3OTM8L2lzYm4+PHVybHM+PHJlbGF0

ZWQtdXJscz48dXJsPmh0dHA6Ly9keC5kb2kub3JnLzEwLjEwNTYvTkVKTW9hMTExMDM1MjwvdXJs

PjwvcmVsYXRlZC11cmxzPjwvdXJscz48ZWxlY3Ryb25pYy1yZXNvdXJjZS1udW0+MTAuMTA1Ni9O

RUpNb2ExMTEwMzUyPC9lbGVjdHJvbmljLXJlc291cmNlLW51bT48YWNjZXNzLWRhdGU+MjAxMi8w

Mi8yMjwvYWNjZXNzLWRhdGU+PC9yZWNvcmQ+PC9DaXRlPjwvRW5kTm90ZT4A

ADDIN EN.CITE.DATA (7) In turn, paraneoplastic thrombocytosis was shown to fuel tumour growth7 and platelet inhibition with aspirin was found to prevent metastatic spread of solid cancers in randomised controlled trials.PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5TdG9uZTwvQXV0aG9yPjxZZWFyPjIwMTI8L1llYXI+PFJl

Y051bT4yMTgzPC9SZWNOdW0+PERpc3BsYXlUZXh0Pig3KTwvRGlzcGxheVRleHQ+PHJlY29yZD48

cmVjLW51bWJlcj4yMTgzPC9yZWMtbnVtYmVyPjxmb3JlaWduLWtleXM+PGtleSBhcHA9IkVOIiBk

Yi1pZD0icHZwYXJ4ejAweHd3MHJlczJwZHhyYXM4NWU5NTJ3ZjUweHdmIj4yMTgzPC9rZXk+PC9m

b3JlaWduLWtleXM+PHJlZi10eXBlIG5hbWU9IkpvdXJuYWwgQXJ0aWNsZSI+MTc8L3JlZi10eXBl

Pjxjb250cmlidXRvcnM+PGF1dGhvcnM+PGF1dGhvcj5TdG9uZSwgUmViZWNjYSBMLjwvYXV0aG9y

PjxhdXRob3I+TmljaywgQWxwYSBNLjwvYXV0aG9yPjxhdXRob3I+TWNOZWlzaCwgSWFpbiBBLjwv

YXV0aG9yPjxhdXRob3I+QmFsa3dpbGwsIEZyYW5jZXM8L2F1dGhvcj48YXV0aG9yPkhhbiwgSGVl

IERvbmc8L2F1dGhvcj48YXV0aG9yPkJvdHRzZm9yZC1NaWxsZXIsIEp1c3RpbjwvYXV0aG9yPjxh

dXRob3I+UnVwYWltb29sZSwgUmFqZXNoYTwvYXV0aG9yPjxhdXRob3I+QXJtYWl6LVBlbmEsIEd1

aWxsZXJtbyBOLjwvYXV0aG9yPjxhdXRob3I+UGVjb3QsIENoYWQgVi48L2F1dGhvcj48YXV0aG9y

PkNvd2FyZCwgSmVybWFpbmU8L2F1dGhvcj48YXV0aG9yPkRlYXZlcnMsIE1pY2hhZWwgVC48L2F1

dGhvcj48YXV0aG9yPlZhc3F1ZXosIEhlcm5hbiBHLjwvYXV0aG9yPjxhdXRob3I+VXJiYXVlciwg

RGlhbmE8L2F1dGhvcj48YXV0aG9yPkxhbmRlbiwgQ2hhcmxlcyBOLjwvYXV0aG9yPjxhdXRob3I+

SHUsIFdlaTwvYXV0aG9yPjxhdXRob3I+R2Vyc2hlbnNvbiwgSGFubmFoPC9hdXRob3I+PGF1dGhv

cj5NYXRzdW8sIEtvamk8L2F1dGhvcj48YXV0aG9yPlNoYWh6YWQsIE1pYW4gTS4gSy48L2F1dGhv

cj48YXV0aG9yPktpbmcsIEVyaW4gUi48L2F1dGhvcj48YXV0aG9yPlRla2VkZXJlbGksIElicmFo

aW08L2F1dGhvcj48YXV0aG9yPk96cG9sYXQsIEJ1bGVudDwvYXV0aG9yPjxhdXRob3I+QWhuLCBF

ZHdhcmQgSC48L2F1dGhvcj48YXV0aG9yPkJvbmQsIFZpcmdpbmlhIEsuPC9hdXRob3I+PGF1dGhv

cj5XYW5nLCBSdWk8L2F1dGhvcj48YXV0aG9yPkRyZXcsIEFuZ2VsYSBGLjwvYXV0aG9yPjxhdXRo

b3I+R3VzaGlrZW4sIEZyYW5jaXNjYTwvYXV0aG9yPjxhdXRob3I+Q29sbGlucywgS2F0aGVyaW5l

PC9hdXRob3I+PGF1dGhvcj5EZUdlZXN0LCBLb2VuPC9hdXRob3I+PGF1dGhvcj5MdXRnZW5kb3Jm

LCBTdXNhbiBLLjwvYXV0aG9yPjxhdXRob3I+Q2hpdSwgV2FoPC9hdXRob3I+PGF1dGhvcj5Mb3Bl

ei1CZXJlc3RlaW4sIEdhYnJpZWw8L2F1dGhvcj48YXV0aG9yPkFmc2hhci1LaGFyZ2hhbiwgVmFo

aWQ8L2F1dGhvcj48YXV0aG9yPlNvb2QsIEFuaWwgSy48L2F1dGhvcj48L2F1dGhvcnM+PC9jb250

cmlidXRvcnM+PHRpdGxlcz48dGl0bGU+UGFyYW5lb3BsYXN0aWMgVGhyb21ib2N5dG9zaXMgaW4g

T3ZhcmlhbiBDYW5jZXI8L3RpdGxlPjxzZWNvbmRhcnktdGl0bGU+TmV3IEVuZ2xhbmQgSm91cm5h

bCBvZiBNZWRpY2luZTwvc2Vjb25kYXJ5LXRpdGxlPjwvdGl0bGVzPjxwZXJpb2RpY2FsPjxmdWxs

LXRpdGxlPk5ldyBFbmdsYW5kIEpvdXJuYWwgb2YgTWVkaWNpbmU8L2Z1bGwtdGl0bGU+PC9wZXJp

b2RpY2FsPjxwYWdlcz42MTAtNjE4PC9wYWdlcz48dm9sdW1lPjM2Njwvdm9sdW1lPjxudW1iZXI+

NzwvbnVtYmVyPjxrZXl3b3Jkcz48a2V5d29yZD5PdmFyaWFuIGNhbmNlcjwva2V5d29yZD48L2tl

eXdvcmRzPjxkYXRlcz48eWVhcj4yMDEyPC95ZWFyPjxwdWItZGF0ZXM+PGRhdGU+MjAxMi8wMi8x

NjwvZGF0ZT48L3B1Yi1kYXRlcz48L2RhdGVzPjxwdWJsaXNoZXI+TWFzc2FjaHVzZXR0cyBNZWRp

Y2FsIFNvY2lldHk8L3B1Ymxpc2hlcj48aXNibj4wMDI4LTQ3OTM8L2lzYm4+PHVybHM+PHJlbGF0

ZWQtdXJscz48dXJsPmh0dHA6Ly9keC5kb2kub3JnLzEwLjEwNTYvTkVKTW9hMTExMDM1MjwvdXJs

PjwvcmVsYXRlZC11cmxzPjwvdXJscz48ZWxlY3Ryb25pYy1yZXNvdXJjZS1udW0+MTAuMTA1Ni9O

RUpNb2ExMTEwMzUyPC9lbGVjdHJvbmljLXJlc291cmNlLW51bT48YWNjZXNzLWRhdGU+MjAxMi8w

Mi8yMjwvYWNjZXNzLWRhdGU+PC9yZWNvcmQ+PC9DaXRlPjwvRW5kTm90ZT4A

ADDIN EN.CITE PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5TdG9uZTwvQXV0aG9yPjxZZWFyPjIwMTI8L1llYXI+PFJl

Y051bT4yMTgzPC9SZWNOdW0+PERpc3BsYXlUZXh0Pig3KTwvRGlzcGxheVRleHQ+PHJlY29yZD48

cmVjLW51bWJlcj4yMTgzPC9yZWMtbnVtYmVyPjxmb3JlaWduLWtleXM+PGtleSBhcHA9IkVOIiBk

Yi1pZD0icHZwYXJ4ejAweHd3MHJlczJwZHhyYXM4NWU5NTJ3ZjUweHdmIj4yMTgzPC9rZXk+PC9m

b3JlaWduLWtleXM+PHJlZi10eXBlIG5hbWU9IkpvdXJuYWwgQXJ0aWNsZSI+MTc8L3JlZi10eXBl

Pjxjb250cmlidXRvcnM+PGF1dGhvcnM+PGF1dGhvcj5TdG9uZSwgUmViZWNjYSBMLjwvYXV0aG9y

PjxhdXRob3I+TmljaywgQWxwYSBNLjwvYXV0aG9yPjxhdXRob3I+TWNOZWlzaCwgSWFpbiBBLjwv

YXV0aG9yPjxhdXRob3I+QmFsa3dpbGwsIEZyYW5jZXM8L2F1dGhvcj48YXV0aG9yPkhhbiwgSGVl

IERvbmc8L2F1dGhvcj48YXV0aG9yPkJvdHRzZm9yZC1NaWxsZXIsIEp1c3RpbjwvYXV0aG9yPjxh

dXRob3I+UnVwYWltb29sZSwgUmFqZXNoYTwvYXV0aG9yPjxhdXRob3I+QXJtYWl6LVBlbmEsIEd1

aWxsZXJtbyBOLjwvYXV0aG9yPjxhdXRob3I+UGVjb3QsIENoYWQgVi48L2F1dGhvcj48YXV0aG9y

PkNvd2FyZCwgSmVybWFpbmU8L2F1dGhvcj48YXV0aG9yPkRlYXZlcnMsIE1pY2hhZWwgVC48L2F1

dGhvcj48YXV0aG9yPlZhc3F1ZXosIEhlcm5hbiBHLjwvYXV0aG9yPjxhdXRob3I+VXJiYXVlciwg

RGlhbmE8L2F1dGhvcj48YXV0aG9yPkxhbmRlbiwgQ2hhcmxlcyBOLjwvYXV0aG9yPjxhdXRob3I+

SHUsIFdlaTwvYXV0aG9yPjxhdXRob3I+R2Vyc2hlbnNvbiwgSGFubmFoPC9hdXRob3I+PGF1dGhv

cj5NYXRzdW8sIEtvamk8L2F1dGhvcj48YXV0aG9yPlNoYWh6YWQsIE1pYW4gTS4gSy48L2F1dGhv

cj48YXV0aG9yPktpbmcsIEVyaW4gUi48L2F1dGhvcj48YXV0aG9yPlRla2VkZXJlbGksIElicmFo

aW08L2F1dGhvcj48YXV0aG9yPk96cG9sYXQsIEJ1bGVudDwvYXV0aG9yPjxhdXRob3I+QWhuLCBF

ZHdhcmQgSC48L2F1dGhvcj48YXV0aG9yPkJvbmQsIFZpcmdpbmlhIEsuPC9hdXRob3I+PGF1dGhv

cj5XYW5nLCBSdWk8L2F1dGhvcj48YXV0aG9yPkRyZXcsIEFuZ2VsYSBGLjwvYXV0aG9yPjxhdXRo

b3I+R3VzaGlrZW4sIEZyYW5jaXNjYTwvYXV0aG9yPjxhdXRob3I+Q29sbGlucywgS2F0aGVyaW5l

PC9hdXRob3I+PGF1dGhvcj5EZUdlZXN0LCBLb2VuPC9hdXRob3I+PGF1dGhvcj5MdXRnZW5kb3Jm

LCBTdXNhbiBLLjwvYXV0aG9yPjxhdXRob3I+Q2hpdSwgV2FoPC9hdXRob3I+PGF1dGhvcj5Mb3Bl

ei1CZXJlc3RlaW4sIEdhYnJpZWw8L2F1dGhvcj48YXV0aG9yPkFmc2hhci1LaGFyZ2hhbiwgVmFo

aWQ8L2F1dGhvcj48YXV0aG9yPlNvb2QsIEFuaWwgSy48L2F1dGhvcj48L2F1dGhvcnM+PC9jb250

cmlidXRvcnM+PHRpdGxlcz48dGl0bGU+UGFyYW5lb3BsYXN0aWMgVGhyb21ib2N5dG9zaXMgaW4g

T3ZhcmlhbiBDYW5jZXI8L3RpdGxlPjxzZWNvbmRhcnktdGl0bGU+TmV3IEVuZ2xhbmQgSm91cm5h

bCBvZiBNZWRpY2luZTwvc2Vjb25kYXJ5LXRpdGxlPjwvdGl0bGVzPjxwZXJpb2RpY2FsPjxmdWxs

LXRpdGxlPk5ldyBFbmdsYW5kIEpvdXJuYWwgb2YgTWVkaWNpbmU8L2Z1bGwtdGl0bGU+PC9wZXJp

b2RpY2FsPjxwYWdlcz42MTAtNjE4PC9wYWdlcz48dm9sdW1lPjM2Njwvdm9sdW1lPjxudW1iZXI+

NzwvbnVtYmVyPjxrZXl3b3Jkcz48a2V5d29yZD5PdmFyaWFuIGNhbmNlcjwva2V5d29yZD48L2tl

eXdvcmRzPjxkYXRlcz48eWVhcj4yMDEyPC95ZWFyPjxwdWItZGF0ZXM+PGRhdGU+MjAxMi8wMi8x

NjwvZGF0ZT48L3B1Yi1kYXRlcz48L2RhdGVzPjxwdWJsaXNoZXI+TWFzc2FjaHVzZXR0cyBNZWRp

Y2FsIFNvY2lldHk8L3B1Ymxpc2hlcj48aXNibj4wMDI4LTQ3OTM8L2lzYm4+PHVybHM+PHJlbGF0

ZWQtdXJscz48dXJsPmh0dHA6Ly9keC5kb2kub3JnLzEwLjEwNTYvTkVKTW9hMTExMDM1MjwvdXJs

PjwvcmVsYXRlZC11cmxzPjwvdXJscz48ZWxlY3Ryb25pYy1yZXNvdXJjZS1udW0+MTAuMTA1Ni9O

RUpNb2ExMTEwMzUyPC9lbGVjdHJvbmljLXJlc291cmNlLW51bT48YWNjZXNzLWRhdGU+MjAxMi8w

Mi8yMjwvYWNjZXNzLWRhdGU+PC9yZWNvcmQ+PC9DaXRlPjwvRW5kTm90ZT4A

ADDIN EN.CITE.DATA (7) ADDIN EN.CITE <EndNote><Cite><Author>Rothwell</Author><Year>2012</Year><RecNum>2201</RecNum><DisplayText>(8)</DisplayText><record><rec-number>2201</rec-number><foreign-keys><key app="EN" db-id="pvparxz00xww0res2pdxras85e952wf50xwf">2201</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Rothwell, PM</author><author>Wilson, M</author><author>Price, JF</author><author>Belch, JF</author><author>Meade, TW</author><author>Mehta, Z</author></authors></contributors><titles><title>Effect of daily aspirin on risk of cancer metastasis: a study of incident cancers during randomised controlled trials. </title><secondary-title>Lancet</secondary-title></titles><periodical><full-title>Lancet</full-title></periodical><pages>Mar 20 [Epub ahead of print]</pages><dates><year>2012</year></dates><urls></urls></record></Cite></EndNote>(8)3.2 Primary care studies. These few studies were set in secondary care. Our team, however, has identified an association between thrombocytosis and several cancers in primary care. ADDIN EN.CITE <EndNote><Cite><Author>Hamilton</Author><Year>2009</Year><RecNum>295</RecNum><DisplayText>(9)</DisplayText><record><rec-number>295</rec-number><foreign-keys><key app="EN" db-id="ratzewttmt2p2oe0e0qpd5zf90w5vvwva52t">295</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Hamilton, W </author></authors></contributors><titles><title>The CAPER studies: five case-control studies aimed at identifying and quantifying the risk of cancer in symptomatic primary care patients</title><secondary-title>British Journal of Cancer</secondary-title></titles><periodical><full-title>British Journal of Cancer</full-title></periodical><pages>S80 – S86</pages><keywords><keyword>Peer-reviewed article</keyword></keywords><dates><year>2009</year></dates><urls></urls></record></Cite></EndNote>(9) In lung cancer, 14% of cases had thrombocytosis, and the positive predictive value (PPV) for lung cancer in a primary care patient with thrombocytosis was 1.6%. ADDIN EN.CITE <EndNote><Cite><Author>Hamilton</Author><Year>2005</Year><RecNum>203</RecNum><DisplayText>(10)</DisplayText><record><rec-number>203</rec-number><foreign-keys><key app="EN" db-id="ratzewttmt2p2oe0e0qpd5zf90w5vvwva52t">203</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Hamilton, W</author><author>Peters, T J</author><author>Round, A</author><author>Sharp, D</author></authors></contributors><titles><title>What are the clinical features of lung cancer before the diagnosis is made? A population based case-control study</title><secondary-title>Thorax</secondary-title><alt-title>Thorax</alt-title></titles><pages>1059-1065</pages><volume>60</volume><keywords><keyword>Peer-reviewed article</keyword></keywords><dates><year>2005</year><pub-dates><date>December 1, 2005</date></pub-dates></dates><urls><related-urls><url>;(10) In colon, the figures were very similar, with a 14% prevalence of thrombocytosis, and a PPV of 1.4% (unpublished data from the CAPER database); ADDIN EN.CITE <EndNote><Cite><Author>Hamilton</Author><Year>2005</Year><RecNum>194</RecNum><DisplayText>(11)</DisplayText><record><rec-number>194</rec-number><foreign-keys><key app="EN" db-id="ratzewttmt2p2oe0e0qpd5zf90w5vvwva52t">194</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Hamilton, W</author><author>Round, A</author><author>Sharp, D.</author><author>Peters, T.</author></authors></contributors><titles><title>Clinical features of colorectal cancer before diagnosis: a population-based case-control study</title><secondary-title>British Journal of Cancer</secondary-title></titles><periodical><full-title>British Journal of Cancer</full-title></periodical><pages>399-405</pages><volume>93</volume><keywords><keyword>Peer-reviewed article</keyword></keywords><dates><year>2005</year></dates><urls></urls></record></Cite></EndNote>(11) in ovary, 12% prevalence and PPV 0.5% (also unpublished reanalysis); ADDIN EN.CITE <EndNote><Cite><Author>Hamilton</Author><Year>2009</Year><RecNum>299</RecNum><DisplayText>(12)</DisplayText><record><rec-number>299</rec-number><foreign-keys><key app="EN" db-id="ratzewttmt2p2oe0e0qpd5zf90w5vvwva52t">299</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Hamilton, William </author><author>Peters, Tim J</author><author>Bankhead, Clare</author><author>Sharp, Deborah</author></authors></contributors><titles><title>Risk of ovarian cancer in women with symptoms in primary care: population based case-control study</title><secondary-title>BMJ 2009</secondary-title></titles><periodical><full-title>BMJ 2009</full-title></periodical><pages>339:b2998</pages><keywords><keyword>Peer-reviewed article</keyword></keywords><dates><year>2009</year></dates><urls></urls></record></Cite></EndNote>(12) similar figures have been identified in our on-going studies of oesophago-gastric (PPV 0.5%), bladder (0.2%) and pancreas cancers (0.2%) (submitted for publication, or in-press). No association was seen in prostate or brain cancers, and the association with uterine cancers was small. ADDIN EN.CITE <EndNote><Cite><Author>Hamilton</Author><Year>2007</Year><RecNum>222</RecNum><DisplayText>(13, 14)</DisplayText><record><rec-number>222</rec-number><foreign-keys><key app="EN" db-id="ratzewttmt2p2oe0e0qpd5zf90w5vvwva52t">222</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Hamilton, W</author><author>Kernick, D</author></authors></contributors><titles><title>Clinical features of primary brain tumours: a case-control study using electronic primary care records.</title><secondary-title>British Journal General Practice</secondary-title></titles><pages>695-699</pages><volume>57</volume><keywords><keyword>Peer-reviewed article</keyword></keywords><dates><year>2007</year></dates><urls></urls></record></Cite><Cite><Author>Hamilton</Author><Year>2006</Year><RecNum>206</RecNum><record><rec-number>206</rec-number><foreign-keys><key app="EN" db-id="ratzewttmt2p2oe0e0qpd5zf90w5vvwva52t">206</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Hamilton, W</author><author>Sharp, D</author><author>Peters, T J</author><author>Round, A</author></authors></contributors><titles><title>Clinical features of prostate cancer before diagnosis: a population-based case-control study</title><secondary-title>British Journal General Practice</secondary-title></titles><pages>756-782</pages><volume>56</volume><keywords><keyword>Peer-reviewed article</keyword></keywords><dates><year>2006</year></dates><urls></urls></record></Cite></EndNote>(13, 14) So, six of the nine primary care cancers we have studied have shown evidence of an association with thrombocytosis. From these six alone, it is clear that the PPV of a cancer in primary care in a patient with thrombocytosis approximates to 5%. Should other cancers show the association (and this seems probable, particularly with breast yet to be studied), then the overall PPV will be even higher. Very few features of cancer in primary care are present in over 10% of cases, and very few features of cancer have a risk of 5% or more, so this is potentially a very useful marker of cancer. ADDIN EN.CITE <EndNote><Cite><Author>Shapley</Author><Year>2010</Year><RecNum>2065</RecNum><DisplayText>(15)</DisplayText><record><rec-number>2065</rec-number><foreign-keys><key app="EN" db-id="pvparxz00xww0res2pdxras85e952wf50xwf">2065</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Shapley, M.</author><author>Mansell, G.</author><author>Jordan, J. L.</author><author>Jordan, K. P.</author></authors></contributors><titles><title>Positive predictive values of &gt;= 5% in primary care for cancer: systematic review</title><secondary-title>British Journal of General Practice</secondary-title></titles><periodical><full-title>British Journal of General Practice</full-title></periodical><pages>681-688</pages><volume>60</volume><number>578</number><keywords><keyword>general Cancer in PC</keyword></keywords><dates><year>2010</year></dates><isbn>0960-1643</isbn><accession-num>WOS:000283572000012</accession-num><urls><related-urls><url>&lt;Go to ISI&gt;://WOS:000283572000012</url></related-urls></urls><electronic-resource-num>10.3399/bjgp10X515412</electronic-resource-num></record></Cite></EndNote>(15) Furthermore, many of these studies preceded routine laboratory transmission of test results to primary care: as a result, there was considerable under-recording of results (both normal and abnormal). In recent years, this has improved. Comparing the data in the published lung paper, which used paper records from 1998-2002 as opposed to the electronic data in the DISCOVERY programme, illustrates this. The ‘old’ figures for lung cancer, as quoted above, were a prevalence of 14% with a PPV of 1.6%. ADDIN EN.CITE <EndNote><Cite><Author>Hamilton</Author><Year>2005</Year><RecNum>203</RecNum><DisplayText>(10)</DisplayText><record><rec-number>203</rec-number><foreign-keys><key app="EN" db-id="ratzewttmt2p2oe0e0qpd5zf90w5vvwva52t">203</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Hamilton, W</author><author>Peters, T J</author><author>Round, A</author><author>Sharp, D</author></authors></contributors><titles><title>What are the clinical features of lung cancer before the diagnosis is made? A population based case-control study</title><secondary-title>Thorax</secondary-title><alt-title>Thorax</alt-title></titles><pages>1059-1065</pages><volume>60</volume><keywords><keyword>Peer-reviewed article</keyword></keywords><dates><year>2005</year><pub-dates><date>December 1, 2005</date></pub-dates></dates><urls><related-urls><url>;(10) The ‘new’ figures are for 5,181 UK lung cancers diagnosed in 2007-9. 3283 (63%) patients with lung cancer had a platelet count in the year before diagnosis. Of these, 343 (10%) were in the range 400-449,000 and 563 (17%) were >450,000. Thus, over a quarter of lung cancers appear to have thrombocytosis before the diagnosis is made. Potentially, the PPV for lung cancer alone is ~3%. While the association of cancer with thrombocytosis is well accepted, raised platelets in a full blood count finding is frequently ignored in both primary and secondary care, because of its low specificity. Furthermore, it is not generally recognised that test results with PPVs of as little even as a few percent can be very valuable in improving the detection rate of early cancer. The need for further studyAll the above suggests the connection between thrombocytosis and cancer is well-researched and understood. On the contrary, this began as an unexpected finding, which is now being replicated in other studies. GPs do not know of the association – and laboratories do not warn of it. If this finding were to be confirmed, it could make a significant contribution to primary care diagnosis of cancer. At the simplest level, it would be possible for laboratories to alert primary care to the possibility of cancer (and provide a list of cancers, ordered in frequency) for any platelet count above an agreed value. More sophisticated solutions (such as programming GP computer systems to alert GPs to thrombocytosis) are also possible. Research questions:4.1 In patients with a raised platelet count in primary care, what is the prevalence of cancer, and which cancers are diagnosed?4.2 What level of platelet count should suggest investigation for cancer? This requires changing the direction of research, so that it is no longer focussed on individual cancers, but on the patient with thrombocytosis. Research planThis is driven by the expected clinical use of the results. We expect them to be of value to a general practitioner receiving a blood count result showing thrombocytosis (probably as an unexpected finding) and wanting to know what (if anything) is to be done in terms of cancer investigation. It is not designed to examine who should have a platelet count measured, or if measurement of a platelet count would be useful as a screening test in an asymptomatic population. Thus, it is a simple cohort study of primary care thrombocytosis. Design: Prospective cohort study using electronic primary care records in the CPRDDetails of the cohort: we will study patients over the age of 40 (this age captures >90% of all new cancers), of either sex, and with a platelet count over 400,000 per mm3. We will only study patients with this count after the beginning of 2000, as electronic submission of blood results to practices became the norm around then. Outcomes: we will identify new diagnoses of cancer occurring in the year after the first raised platelet count. This will be done by searching the clinical/referral files of patients within the cohort for medcodes representing cancer. Medcodes are the main variable within the CPRD used to store diagnostic information. Of the 100,035 possible medcodes, 2,623 are cancer diagnostic codes, which the PI on this study has categorised into the 18 main adult cancer sites, plus a 19th miscellaneous category. This list has been adopted by the CPRD for its cancer studies. One year has been chosen as a reasonable compromise between identifying all cancers related to the thrombocytosis, yet reducing the chance of identifying unrelated cancers. The results will be presented as the rate of cancer in patients with thrombocytosis (an incidence figure). Sub-analyses will examineThe type of cancers occurringThe incidence of cancer in each gender group amongst those with thrombocytosisThe incidence of cancer in different age groups (probably by 10 year age bands) amongst those with thrombocytosis Different levels of thrombocytosis (probably 400-424, 425-449, 450+). There is sufficient power (see below) to present this as a ROC curve of cancer incidence vs. platelet count. The ROC analysis should allow us to identify the optimal cut-point level of thrombocytosis for discriminating between those who do and do not develop cancer.We will also request the dataset to be linked to cancer registries for confirmation of cancer diagnoses (especially the date of diagnosis). Staging data is patchily present in linked registry-CPRD datasets and our previous experience suggests it may not be possible to perform a reliable sub-analysis of stage. Sample size/power calculations. The CPRD prefer to limit their datasets to 50,000 patients. The prevalence of thrombocytosis is such that they will easily be able to provide this number, and we propose to request 40,000 with thrombocytosis (Gallagher: personal communication). We have performed three sample size calculations based on estimating the 12 month cancer incidence amongst those with thrombocytosis. To estimate a cancer incidence of 5% with an error margin of ±2% requires 457 subjects with thrombocytosis.To estimate a cancer incidence of 10% prevalence with an error margin of ±2% requires 865 subjects with thrombocytosis.To estimate the area under the curve of 80% with an error margin of ±1% in an ROC curve analysis of cancer incidence (binary reference standard) by platelet count (continuous test) requires 38,254 subjects, assuming the ratio of cancers to non-cancers is 1:19 (i.e. a 5% prevalence of cancer). Note that these subjects are not selected with respect to thrombocytosis status since we need both subjects with and without thrombocytosis to do this analysis. ADDIN EN.CITE <EndNote><Cite><Author>Zhou</Author><Year>2002</Year><RecNum>629</RecNum><DisplayText>(16)</DisplayText><record><rec-number>629</rec-number><foreign-keys><key app="EN" db-id="pvparxz00xww0res2pdxras85e952wf50xwf">629</key></foreign-keys><ref-type name="Book">6</ref-type><contributors><authors><author>Zhou, X-H</author><author>Obuchowski, NA</author><author>McClish, DK</author></authors></contributors><titles><title>Statistical methods in Diagnostic medicine</title></titles><keywords><keyword>Methods</keyword></keywords><dates><year>2002</year></dates><pub-location>New York</pub-location><publisher>Wiley</publisher><urls></urls></record></Cite></EndNote>(16)Thus the proposed dataset of 40,000 patients has sufficient power to answer the main research questions, and to estimate cancer incidence precisely within subgroups of thrombocytosis patients defined by age and sex, providing each subgroup is sized around 500 to 1000. We will request an age-sex matched sample of 10,000 patients (selected at random to match a quarter of the 40,000) who have had a normal platelet count measure. Although not needed for the main outcome (PPV), this comparison sample will allow us to determine the completeness of recording of cancer diagnoses within the study, and will address any concerns that there could be confounding by indication (whereby more ‘ill’ patients are more likely to receive a blood test, and are thus more likely to have thrombocytosis identified. This comparison sample has over 99% power to identify a change in incidence of cancer from 2% in the comparison group to 3% in the thrombocytosis group (we expect it to be higher).The study team and environment. This consists of a professor of primary care diagnostics (Hamilton), an experienced methodologist/statistician (Okoumunne), a professor of diagnostics (Hyde) and a consultant haematologist (Rudin). It will be hosted within the DISCOVERY research group of the University of Exeter Medical School. This research group is highly experienced in cancer diagnostic studies, and in using CPRD data. The environment is also ideal for any implementation of the findings, as the Medical School and NHS units make up the Peninsula CLAHRC, tasked with implementation research – particularly embedding research findings into clinical practice. Outputs, dissemination and implementation plan. Conventional outputs will include national and international conference presentations and high profile peer reviewed publications. We have strong links with the Department of Health, the National Cancer Action Team and several cancer charities. They will be fed results in a format most useful to them. Each of these three bodies is moving towards integration of cancer diagnostics into GP computers, and it would be relatively easy to add any thrombocytosis-related action into current initiatives. 8. References1.M. P. Coleman, D. Forman, H. Bryant, J. Butler, B. Rachet, C. Maringe, U. Nur, E. Tracey, M. Coory, J. Hatcher, C. E. McGahan, D. Turner, L. Marrett, M. L. Gjerstorff, T. B. Johannesen, J. Adolfsson, M. Lambe, G. Lawrence, D. Meechan, E. J. Morris, R. Middleton, J. Steward and M. A. Richards. Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995-2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data. The Lancet. 2011;377(9760):127-38.2.Hamilton W. Cancer diagnosis in primary care. British Journal of General Practice. 2010 Feb;60(571):121-8.3.NICE. Referral Guidelines for suspected cancer. London: NICE; 2005.4.Hamilton W. Computer assisted diagnosis of ovarian cancer in primary care. BMJ. 2012;344:10.1136/bmj.d7628.5.Levin J, Conley C. Thrombocytosis associated with malignant disease. Archives of Internal Medicine. 1964;114:497-500.6.Costantini V, Zacharski LR, Moritz TE, Edwards RL. The platelet count in carcinoma of the lung and colon. Thrombosis & Haemostasis. 1990;64(4):501-5.7.R. L. Stone, A. M. Nick, I. A. McNeish, F. Balkwill, H. D. Han, J. Bottsford-Miller, R. Rupaimoole, G. N. Armaiz-Pena, C. V. Pecot, J. Coward, M. T. Deavers, H. G. Vasquez, D. Urbauer, C. N. Landen, W. Hu, H. Gershenson, K. Matsuo, M. M. K. Shahzad, E. R. King, I. Tekedereli, B. Ozpolat, E. H. Ahn, V. K. Bond, R. Wang, A. F. Drew, F. Gushiken, K. Collins, K. DeGeest, S. K. Lutgendorf, W. Chiu, G. Lopez-Berestein, V. Afshar-Kharghan and A. K. Sood. Paraneoplastic Thrombocytosis in Ovarian Cancer. New England Journal of Medicine. 2012 2012/02/16;366(7):610-8.8.Rothwell P, Wilson M, Price J, Belch J, Meade T, Mehta Z. Effect of daily aspirin on risk of cancer metastasis: a study of incident cancers during randomised controlled trials. . Lancet. 2012:Mar 20 [Epub ahead of print].9.Hamilton W. The CAPER studies: five case-control studies aimed at identifying and quantifying the risk of cancer in symptomatic primary care patients. British Journal of Cancer. 2009:S80 – S6.10.Hamilton W, Peters TJ, Round A, Sharp D. What are the clinical features of lung cancer before the diagnosis is made? A population based case-control study. Thorax. 2005 December 1, 2005;60:1059-65.11.Hamilton W, Round A, Sharp D, Peters T. Clinical features of colorectal cancer before diagnosis: a population-based case-control study. British Journal of Cancer. 2005;93:399-405.12.Hamilton W, Peters TJ, Bankhead C, Sharp D. Risk of ovarian cancer in women with symptoms in primary care: population based case-control study. BMJ 2009. 2009:339:b2998.13.Hamilton W, Kernick D. Clinical features of primary brain tumours: a case-control study using electronic primary care records. British Journal General Practice. 2007;57:695-9.14.Hamilton W, Sharp D, Peters TJ, Round A. Clinical features of prostate cancer before diagnosis: a population-based case-control study. British Journal General Practice. 2006;56:756-82.15.Shapley M, Mansell G, Jordan JL, Jordan KP. Positive predictive values of >= 5% in primary care for cancer: systematic review. British Journal of General Practice. 2010;60(578):681-8.16.Zhou X-H, Obuchowski N, McClish D. Statistical methods in Diagnostic medicine. New York: Wiley; 2002.17.Tate AR, Martin AGR, Ali A, Cassell JA. Using free text information to explore how and when GPs code a diagnosis of ovarian cancer: an observational study using primary care records of patients with ovarian cancer. BMJ Open. 2011. ADDIN EN.REFLIST 1.Coleman MP, Forman D, Bryant H, Butler J, Rachet B, Maringe C, et al. Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995-2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data. The Lancet. 2011;377(9760):127-38.2.Hamilton W. Cancer diagnosis in primary care. British Journal of General Practice. 2010 Feb;60(571):121-8.3.NICE. Referral Guidelines for suspected cancer. London: NICE; 2005.4.Hamilton W. Computer assisted diagnosis of ovarian cancer in primary care. BMJ. 2012;344:10.1136/bmj.d7628.5.Levin J, Conley C. Thrombocytosis associated with malignant disease. Archives of Internal Medicine. 1964;114:497-500.6.Costantini V, Zacharski LR, Moritz TE, Edwards RL. The platelet count in carcinoma of the lung and colon. Thrombosis & Haemostasis. 1990;64(4):501-5.7.Stone RL, Nick AM, McNeish IA, Balkwill F, Han HD, Bottsford-Miller J, et al. Paraneoplastic Thrombocytosis in Ovarian Cancer. New England Journal of Medicine. 2012 2012/02/16;366(7):610-8.8.Rothwell P, Wilson M, Price J, Belch J, Meade T, Mehta Z. Effect of daily aspirin on risk of cancer metastasis: a study of incident cancers during randomised controlled trials. . Lancet. 2012:Mar 20 [Epub ahead of print].9.Hamilton W. The CAPER studies: five case-control studies aimed at identifying and quantifying the risk of cancer in symptomatic primary care patients. British Journal of Cancer. 2009:S80 – S6.10.Hamilton W, Peters TJ, Round A, Sharp D. What are the clinical features of lung cancer before the diagnosis is made? A population based case-control study. Thorax. 2005 December 1, 2005;60:1059-65.11.Hamilton W, Round A, Sharp D, Peters T. Clinical features of colorectal cancer before diagnosis: a population-based case-control study. British Journal of Cancer. 2005;93:399-405.12.Hamilton W, Peters TJ, Bankhead C, Sharp D. Risk of ovarian cancer in women with symptoms in primary care: population based case-control study. BMJ 2009. 2009:339:b2998.13.Hamilton W, Kernick D. Clinical features of primary brain tumours: a case-control study using electronic primary care records. British Journal General Practice. 2007;57:695-9.14.Hamilton W, Sharp D, Peters TJ, Round A. Clinical features of prostate cancer before diagnosis: a population-based case-control study. British Journal General Practice. 2006;56:756-82.15.Shapley M, Mansell G, Jordan JL, Jordan KP. Positive predictive values of >= 5% in primary care for cancer: systematic review. British Journal of General Practice. 2010;60(578):681-8.16.Zhou X-H, Obuchowski N, McClish D. Statistical methods in Diagnostic medicine. New York: Wiley; 2002. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download