This proposal is to fund a pilot study to assess feasibility and refine methodology for an intended large Scotland wide study on Response to Oral Agents in Diabetes (ROAD). The study will collect cohorts of patients who have carefully controlled standardised dose titration and monitoring with an assessment of drug response and side effects over a 6 month period. The primary aim will be to use these cohorts to investigate phenotypic and genotypic (pharmacogenetic) determinants of response.

Drug naïve patients will be treated with Metformin. Patients who have failed on Metformin or are intolerant of Metformin will be randomised to gliclazide, pioglitazone or sitagliptin. With the ability to capture patient data beyond 6 months via data linkage we will monitor time to treatment failure and therefore compare which of the 3 oral agents is the best therapy to use after Metformin in a cost efficient and "real world" RCT.

The Response to Oral Agents in Diabetes (ROAD) study aims to address the limitations of observational data by creating a prospective study of incident users of oral agents. For the first six months the research team will ensure a protocol driven dose titration, standardised monitoring of adherence, response and side effects and standardised deviations from therapy. Thereafter patients will receive 6 monthly monitoring and further protocol led dose titration by the GP. Biochemistry, prescribing data, morbidity and mortality data will be captured for up to 10 years from drug initiation. The ROAD study will provide a highly powered prospective cohort to investigate phenotypic and genotypic determinants of response in its own right. However, this cohort will be used synergistically with ongoing observational pharmacogenetics studies, allowing for crucial replication of 'positive' signals. Furthermore, by randomisation at drug initiation, long term community follow up will allow a comparison of time to treatment failure in patients treated with gliclazide, pioglitazone and sitagliptin in a much more cost effective and 'real-world' setting than traditional prospective randomised trials This pilot study is to assess the feasibility of the larger complex intervention. The primary outcome of the pilot is HbA1c change. Other measures regarding recruitment and dose titration will be assessed. With knowledge from this pilot, an application will be made for a large region or Scotland wide study to collect 2000 patients incident to oral diabetes treatment.

Inclusion Criteria:

• Cohort 1 - metformin treatment

  • Type 2 diabetes diagnosed more than 6 weeks prior to visit 1
  • GP considers adequate diet and lifestyle advice given
  • Age >35 and < 80
  • Age of diabetes diagnosis >35
  • White European
  • HbA1c >7% & <=9%
  • eGFR>=50 ml/min
  • ALT <= 2.5*ULN
  • Contactable by telephone

• Cohort 2 - 2nd line treatment

  • Type 2 diabetes
  • Treated with metformin for more than 3 months; or metformin intolerant
  • Age >35 and < 80
  • Age of diabetes diagnosis >35
  • White European
  • HbA1c >7% & <=9%
  • eGFR>=50 ml/min
  • ALT <= 2.5*ULN
  • No previous history of heart failure; No patients with documented evidence of left ventricular systolic dysfunction OR with symptoms and signs consistent with a clinical diagnosis of heart failure
  • No treatment with Gemfibrozil or Rifampicin (CYP2C8 inhibitor or inducer respectively); or with Miconazole or phenylbutazone (increased hypoglycemic effect of gliclazide).
  • No diagnosis of osteoporosis
  • Contactable by telephone

Exclusion Criteria:

• Cohort 1

  • Type 1 diabetes
  • HbA1c >9% or <=7%
  • eGFR<50 ml/min
  • ALT > 2.5*ULN
  • Alcohol consumption in excess of 50 units per week
  • Pregnancy, lactation or a female planning to conceive within the study period
  • Any other significant medical reason for exclusion as determined by the investigator

• Cohort 2

  • Type 1 diabetes
  • HbA1c >9% or <=7%
  • eGFR< 50 ml/min
  • ALT > 2.5*ULN
  • Previous history of heart failure OR documented evidence of left ventricular systolic dysfunction OR symptoms and signs consistent with a clinical diagnosis of heart failure
  • Ongoing treatment with Gemfibrozil or Rifampicin (CYP2C8 inhibitor or inducer respectively); or with Miconazole or phenylbutazone (increased hypoglycemic effect of gliclazide).
  • Previous diagnosis of osteoporosis
  • Pregnancy, lactation or a female planning to conceive within the study period
  • Any other significant medical reason for exclusion as determined by the investigator

Note that this study was a pilot study and was not designed or powered to show significance of differences in outcomes between groups. Primary aim was to demonstrate efficacy of novel route of recruitment from primary care.