Scientific method hypothesis formulation mentorship

1
Scientific methodhypothesis formulation
mentorship
GCRC training course 2006
Philip A. Kern, M.D. GCRC Associate Director
2
Scientific Method

• Ask a question
• Pose a hypothesis
• Ok, now prove it

3
What is a hypothesis?

• Dictionary an unproved theory, proposition,
  tentatively accepted to explain certain facts or
  to provide a basis for further investigation

4
Hypothesis-driven research

• Not all research is hypothesis driven
• Some pharmaceutical research involves testing a
  drug for safety and effectiveness vis a vis
  standard therapy
• Some studies are necessarily descriptive.
  Hypothesis testing cannot begin until the problem
  can be better described
• Some studies are glorious fishing expeditions,
  eg. Microarray studies. Such studies may provide
  the basis for later hypothesis testing

5
Examples from the past

• Type 1 diabetes
• Familial hypercholesterolemia
• Leptin
• Obesity
• Lipodystrophy

6
A pancreatic protein extract, when injected into
an individual with diabetes, will lower blood
glucose and reverse the diabetic condition
Examples from the past
Type 1 diabetes (c. 1920). Children would
develop ketoacidosis. They could be resuscitated
temporarily with fluids, carbohydrate
restriction. But they would inevitably become
cachectic and die.
7
The discovery of insulin

• 1921 Discovery of insulin
• Frederick Banting, MD, a surgeon in London,
  Ontario
• Charles Best, a medical student

8
Familial Hypercholesterolemia and the LDL receptor

• A 38 yo man in otherwise good health stops by
  the side of the road coming home from work to
  vomit. Chest pressure overwhelms him, and he is
  DOA on arrival at the hospital. His cholesterol
  is 400 mg/dl (heterozygote).
• A mother takes her 8 yo daughter to the doctor
  because of chest pain and dyspnea when she plays
  soccer. The astute pediatrician does blood
  tests, an EKG and CXR. The child has acute
  coronary ischemia, cardiomegaly, and her
  cholesterol is 950 mg/dl (homozygote)

9
The LDL Receptor Story
10
Another example of hypothesis testing from the
1980sHypothesis

• In a child with LDL receptor deficiency, a liver
  transplant from a normal subject will provide
  needed LDL receptors, resulting in lower LDL
  cholesterol and reverse of atherosclerosis

11
A 6 year old with myocardial infarction, hyper-
cholesterolemia, and absent LDL receptors
Recently, a more direct therapeutic approach was
taken in an FH homozygote, whose initials are
S.J. and who has two mutant genes at the LDL
receptor locus. This six-year-old girl had a
total plasma cholesterol level over 1,000 mg/dl
and she sustained repeated episodes of myocardial
infarction. After she failed to respond to two
coronary bypass procedures plus a mitral valve
replacement, she was subjected to combined
heart-liver transplantation by a team of surgeons
led by Thomas E. Starzl at the University of
Pittsburgh. The liver transplant was designed to
provide a source of LDL receptors. Immediately
after the operation, S.J.s total plasma
cholesterol level fell from 1100 mg/dl to the
range of 200 to 300 mg/dl, and it remained in
that range for the succeeding 13 months.
12
Fig. 25. LDL metabolism in S.J., a patient with
homozygous FH, before and after liver-heart
transplantation. Panel A Total cholesterol
levels in plasma. Panel B Plasma decay curves of
1251-LDL after intravenous injection of tracer
amounts of 125I-LDL before (A) and after (A)
liver-heart transplantation. (Data in Panel B
reprinted with permission from ref. 115).
A RECEPTOR-MEDIATED PATHWAY FOR CHOLESTEROL
HOMEOSTASIS Nobel lecture, 9 December,
1985 by MICHAEL S. BROWN AND JOSEPH L.
GOLDSTEIN Department of Molecular Genetics,
University of Texas Health Science Center,
Southwestern Medical School, 5323 Harry Hines
Blvd. Dallas, Texas, U.S.A.
13
Obesity The obese mouse and littermate
14
Leptin and the ob/ob mouse

• Defect is a point mutation in leptin. Leptin is
  expressed by adipose tissue, and signals the
  hypothalamus to stop feeding. Leptin also
  stimulates energy utilization.
• Restoring leptin inhibits food consumption and
  normalizes the phenotype

The ob/ob mouse obese, hyperphagic, low
metabolic rate, defective thermogenesis,
diabetic, insulin resistant, infertile
15
Hypothesis

• Leptin deficiency in rodents leads to obesity.
  Therefore, human obesity is due to a deficiency
  in leptin production by adipose tissue, leading
  to a failure to suppress appetite,
  overconsumption of food, and obesity.

16
Plasma leptin is elevated in obese humans
Hypothesis of leptin deficiency causing obesity
is disproved
Obese
Impaired fertility
Overweight
Thin
17
Obesity and DiabetesThe paradox of lipodystrophy

• Obesity excess adipose tissue leading to
  abnormal fat deposition, insulin resistance and
  type 2 diabetes
• Lipodystrophy absence of adipose tissue leading
  to abnormal fat deposition and insulin resistance
• Fatty liver is a key component of both condtions

18
Mouse lipodystrophy model

• Essentially absent adipose tissue
• Diabetes/insulin resistance
• Fatty liver and muscle
• Hypertriglyceridemia

• Decreased liver lipid
• Decreased muscle lipid

A-ZIP/F dominant negative binds B-Zip
transcription factors adipose specific JCI
106 1221, 2000
19
Treatment of lipodystrophic mice with a fat
transplant

• Decreased liver lipid
• Decreased muscle lipid

20
Congenital lipodystrophy in humans

• Almost no body fat (appearance of muscularity)
• Extreme insulin resistance and diabetes
• Ectopic lipid accumulation (large fatty livers)
• Hypertriglyceridemia

21
Hypothesis

• A secretory product of adipose tissue is absent
  in patients with lipodystrophy
• This secretory product of adipose tissue is
  necessary for normal lipid metabolism in other
  tissues
• This secretory product is leptin

22
Treatment of human lipodystrophy with leptin
Oral et al, NEJM 346570, 2002
23
Treatment of human lipodystrophy with leptin
Oral et al, NEJM 346570, 2002
24
Have all the good experiments already been done?
1921
1985
2002
25
Scientific method and investigator initiated
grants

• Introduction
• Hypotheses
• Specific aims
• Background and significance
• Critical questions remaining to be answered
• Preliminary data
• Methods
• Human subjects/DSMP
• References cited

26
Scientific method and investigator initiated
grants

• Introduction
• Hypotheses
• Specific aims

• This is where you lay out the big picture and
  generate excitement for the project
• Introduction.
• It is important to state the nature of the
  problem in enough detail to make the hypotheses
  and aims understandable.
• This is the first section the reviewers will
  read. Good first impressions last.
• Many members of the study section may read only
  this part of the grant

27
Scientific method and investigator initiated
grants

• Specific aims/hypotheses
• Pose a question/state a hypothesis (eg. A
  pancreatic extract contains a substance which
  reverses the diabetic condition)
• State an aim (eg. Subjects with diabetes will be
  randomized to receive injections of either a
  pancreatic endocrine extract or placebo)
• Do not simply state what you will measure
• We will measure urine sugar in diabetic subjects
  given pancreatic extracts

28
Scientific method and investigator initiated
grants

• Introduction
• Hypotheses
• Specific aims
• Background and significance
• Critical questions remaining to be answered
• Preliminary data
• Methods
• Human subjects/DSMP
• References cited

29
Scientific method and investigator initiated
grants

• Background and significance
• Critical questions remaining to be answered

• Tell a story
• Begin with fundamental concepts
• Diabetes is a horrible untreatable disease
• Removing the pancreas of a dog causes a syndrome
  resembling human diabetes
• Provide the reader with the information (s)he
  needs to understand your grant
• Understand your reviewers
• Finish the story
• Important knowledge gaps
• Important questions remaining to be answered
• Hopefully, these will be your specific aims

30
Scientific method and investigator initiated
grants

• Introduction
• Hypotheses
• Specific aims
• Background and significance
• Critical questions remaining to be answered
• Preliminary data
• Methods
• Human subjects/DSMP
• References cited

31
Scientific method and investigator initiated
grants

• Preliminary data
• Are there critical methodologic hurdles that you
  need to overcome?
• Pancreatic tissue contains lipases and
  proteases, and we developed a method for removing
  them so the extract will not cause tissue
  necrosis
• Can you provide evidence that the experiments are
  feasible?
• We already recruited 15 subjects, which required
  screening of 45 subjects..
• Proof of concept
• We performed these experiments in a
  pancreatecomized dog, and demonstrated lowered
  blood glucose, decrease in ketones.
• Figures/tables can be extremely valuable

32
Scientific method and investigator initiated
grants

• Introduction
• Hypotheses
• Specific aims
• Background and significance
• Critical questions remaining to be answered
• Preliminary data
• Methods
• Human subjects/DSMP
• References cited

33
Scientific method and investigator initiated
grants

• Methods
• Restate your aims/hypotheses
• Write your methods such that your methods answer
  your aims/hypotheses.
• Do not assume that the reviewer will understand
  how your methods answer your aims
• Clinical trial/recruitment
• Who is being recruited
• Inclusion/exclusion criteria-be complete
• Table/figure/study design helpful
• Potential experimental problems
• Sample size calculation
• Need not necessarily be lengthy, but should be
  reasoned and evidenced-based
• Future directions
• Sometimes a place for risky experiments

34
Mentorship
35
Selecting a mentor

• Expertise
• Scientific
• Clinical
• Methodologic
• Conceptual
• Has been through the process
• Currently funded
• Sits on review panels
• Time
• Will read your grant
• Do you sense an interest, even excitement?

36
Expectations

• Mentor
• Considers your work important maybe even gets
  excited by it
• Ideas take root and grow
• Reads your grant and provides critique
• Direct you to other resources
• Expect conceptual guidance more than technical
  detail
• Acknowledge limitations
• Mentee
• Works hard
• Reads the literature
• Teaches the mentor something
• Does not expect spoon-feeding
• Tries to discover solutions and solve problems
• Honesty

37
(No Transcript)