Chapter 11 — Further Reading
Tier 1 — Verified canonical
The discovery. The Nobel Prize in Physiology or Medicine 1923 was awarded to Frederick Banting and J.J.R. Macleod for the discovery of insulin. Banting shared his portion with Charles Best; Macleod shared his with James Collip. The Nobel organization publishes the lectures and biographical material free, and Banting's lecture is worth reading — it is unusually clear-eyed about what insulin is and is not.
The University of Toronto Libraries maintain a digitized collection of the discovery-era papers, notebooks, and correspondence. It is freely accessible and remarkable: you can read the laboratory notebooks.
Insulin structure. Frederick Sanger's determination of insulin's amino acid sequence (Nobel Prize in Chemistry 1958) is covered in Chapter 1's Case Study 1. Dorothy Hodgkin determined insulin's three-dimensional structure by X-ray crystallography, work for which she had already received the Nobel Prize in Chemistry in 1964 for other structures; the insulin structure came later and is the basis for understanding the hexamer.
Recombinant human insulin was approved in 1982 — the first recombinant DNA drug approved anywhere. Approval records are public via Drugs@FDA.
The DCCT. The Diabetes Control and Complications Trial was publicly funded by what is now the National Institute of Diabetes and Digestive and Kidney Diseases, reported in 1993, and demonstrated that intensive glycemic control reduces microvascular complications in type 1 diabetes — while producing more severe hypoglycemia. Its observational extension, EDIC, followed participants afterward. NIDDK publishes accessible summaries of both.
Labels. DailyMed carries current labeling for every insulin product. Comparing the labels for a rapid-acting analog, NPH, and a long-acting analog side by side is the clearest way to see the timing problem §11.6 describes — the onset, peak, and duration figures are stated plainly.
Closed-loop systems. Automated insulin delivery systems have been cleared by regulators on the basis of randomized trials; device clearance records are public via the FDA device databases.
Tier 2 — Attributed, specifics unverified
On the discovery timeline. The account in §11.1 is compressed. The precise contributions of Banting, Best, Macleod, and Collip have been the subject of longstanding historical dispute, and the Nobel award caused lasting acrimony. This book states the outline and does not adjudicate.
On animal insulin. Porcine insulin differs from human insulin at one residue and bovine at three. Animal-sourced insulin was in clinical use for roughly six decades and was associated with variable purity and with anti-insulin antibody formation.
On analog pharmacokinetics. Onset, peak, and duration figures for each analog category are stated in product labels and vary by product, dose, and individual. This book describes the strategies rather than quoting times, deliberately.
On analogs versus human insulin. Reduced nocturnal and severe hypoglycemia with analogs in type 1 diabetes is well supported. The incremental benefit in type 2 diabetes is smaller and has been genuinely contested, which matters given the cost difference.
On closed-loop systems. Randomized trials have reported increased time in target glucose range and reduced hypoglycemia across age groups. Systems differ substantially, and comparative data between systems is thinner than data against conventional therapy.
On metabolic memory. The EDIC observational extension reported persistent advantages in the originally intensively treated group after glycemic differences narrowed. The phenomenon is widely described; proposed mechanisms including epigenetic modification and accumulated tissue changes are not established. The extension is observational, with the limitations Chapter 10's Case Study 2 catalogues.
On insulin pricing and rationing. That U.S. list prices rose substantially over recent decades, that prices exceed those in most other high-income countries for the same products, and that a meaningful minority of people with diabetes report cost-related underuse are all well documented. Reported proportions vary substantially by population, insurance status, and survey method, and this book quotes no figure. Individual deaths following insulin rationing have been documented and publicly investigated. Out-of-pocket cap programs, direct-purchase initiatives, and biosimilar entry have reduced costs for many people in recent years; any specific current figure would date quickly.
Tier 3 — Illustrative and constructed
- The insulin structure diagram, the signaling diagram, the two-lever counter-regulation figure, the analog strategy table, and Case Study 2's five-feature diagram are this book's own teaching devices.
- The worked Field 5-at-depth dossier entry for insulin is a demonstration constructed for this book.
- Figures 11.1 and 11.CS1 render the real historical record and a real published trial in this book's six-field format; the format is the book's own.
If you only do one thing
Read the University of Toronto's digitized discovery collection for twenty minutes.
The notebooks are there. You can see the dog numbers, the extract preparations, the glucose readings. It is the origin of every peptide drug in this book, and it looks like what it was: a small group of people in a poorly funded laboratory, arguing, making a mess, and getting it right.
If you would rather read one modern document: open three insulin labels on DailyMed — a rapid-acting analog, NPH, and a long-acting analog — and compare their onset, peak, and duration. The whole of §11.6 is in those three numbers, and seeing them side by side makes the engineering problem obvious in a way that prose does not.
Looking ahead
Chapter 12 covers cost, access, shortage, compounding, and the argument about what obesity is.
Useful preparation: re-read this chapter's Case Study 2, particularly the five-feature diagram, and write down which features you expect the GLP-1 drugs to share. Chapter 12 will tell you whether you were right, and the prediction is more instructive than the answer.