Eli Lily and Company: Innovation in Diabetes Case analysis All requirements are in the attached documents. For the exclusive use of d. feng, 2019. 9-696-07

Eli Lily and Company: Innovation in Diabetes Case analysis All requirements are in the attached documents. For the exclusive use of d. feng, 2019.
9-696-077
REV: APRIL 15, 2004
CLAYTON M. CHRISTENSEN
Eli Lilly and Company: Innovation in Diabetes Care
“Look at these. Aren’t they beautiful?” asked Larry Ellingson, executive director of Eli Lilly and
Company’s Diabetes Care Business Unit, as he showed his visitor a briefcase full of odd looking
plastic devices. “They’re pens—insulin pens. All you do, “ he explained as he screwed one apart, “is
put a little cartridge of insulin in here; close it up like this; turn this dial to the amount of insulin you
need; poke the needle just under your skin (which he didn’t demonstrate); and squeeze this trigger.
That’s all there is to it. Then you just put this cap over the needle and put it back in your briefcase,
purse, or pocket until your next meal, when you take a shot again. The patients will just love it.”
If history was any guide, Ellingson was right: they would love it. Lilly’s principal competitor in
the worldwide insulin business, Denmark-based Novo Nordisk, had introduced insulin pens to the
European market several years earlier with great success. Pens were a more convenient way for
patients to take insulin. Conventionally, patients carried a separate syringe, inserted its needle into
an insulin vial, pulled its plunger out to draw slightly more than the desired amount of insulin into
the syringe, flicked the syringe while holding its needle up to dislodge any air bubbles that clung to
the walls of the syringe’s cylinder, and then squeezed the plunger slightly to force those bubbles—
and some insulin—out of the syringe. Only then could they inject themselves with insulin. This
process typically took more than a minute, whereas patients could prepare and administer a pen
injection in as little as 10 seconds.
It was early in 1995, and Novo was building a new plant in the United States to produce insulin
cartridges for its pens. Ellingson hoped that Lilly’s new line of pens (see Exhibit 1), the result of a
multimillion dollar investment, would blunt the advantage Novo had enjoyed with convenienceconscious customers and stabilize Lilly’s share of the worldwide insulin market.
Insulin was an important product for Lilly, one of the world’s largest pharmaceutical
manufacturers with sales of over $5 billion (see Exhibit 2). Insulin in fact was Lilly’s second-largest
revenue producer after its widely prescribed drug for depression, Prozac.
Diabetes and Insulin
Diabetes is actually two fundamentally different diseases that share a similar set of symptoms:
Type I patients produce no insulin, the hormone necessary for cells to utilize glucose, while Type II
patients cannot efficiently use the insulin their bodies produce. Type I, also known as juvenile
diabetes, usually begins during childhood or puberty. Type II, known as adult-onset diabetes, is
manifest later in life (usually after the age of 40) and usually is associated with—and possibly caused
________________________________________________________________________________________________________________
Professor Clayton M. Christensen prepared this case. HBS cases are developed solely as the basis for class discussion. Cases are not intended to
serve as endorsements, sources of primary data, or illustrations of effective or ineffective management.
Copyright © 1996 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800-545-7685,
write Harvard Business School Publishing, Boston, MA 02163, or go to http://www.hbsp.harvard.edu. No part of this publication may be
reproduced, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means—electronic, mechanical,
photocopying, recording, or otherwise—without the permission of Harvard Business School.
This document is authorized for use only by di feng in SPRING 2019 (UNGER) THE INNOVATION PROCESS-1-1 taught by BARRY UNGER, Boston University from Jan 2019 to May 2019.
For the exclusive use of d. feng, 2019.
696-077
Eli Lilly and Company: Innovation in Diabetes Care
by—obesity. Digestive processes convert most food into glucose (a simple sugar) and then pass that
glucose into the blood as the body’s main source of energy. Body cells are able to burn or metabolize
glucose, however, only when there is insulin present, acting as a sort of catalyst for “burning” the
glucose. Because they either cannot produce insulin in the pancreas (Type I) or use the insulin they
produce (Type II), those with diabetes (hereafter simply called “patients”) can have high
concentrations of unmetabolizable glucose in their bloodstream.
Patients need to inject the precise amount of insulin required to metabolize the glucose produced
by their digestive system. If they inject too little, the resultant high blood-sugar levels cause a slow
deterioration of the body, particularly of the eyes and kidneys. Low blood-sugar levels caused by an
overdose of insulin, though, can rapidly precipitate unconsciousness and, potentially, death.
Many Type II patients can treat their condition with oral medications that either cause their
pancreases to produce more insulin or enhance the sensitivity of their body tissues to the insulin they
naturally produce. Some Type II and all Type I patients, however, must take daily injections of
insulin to survive. Insulin cannot be taken orally because it is a protein and would be broken down
by the digestive system.
All Type I patients must begin insulin injections immediately upon diagnosis. Most Type II
patients, however, progress through several stages. Typically, their diabetes initially is so mild (they
can use most of the insulin they produce) that they aren’t symptomatic1—their illness remains
undiagnosed until it is discovered during a routine physical exam or in the course of treatment for
some other disease. Upon initial diagnosis, many can reduce their levels of blood glucose to normal
through a combination of diet, exercise, and weight loss. Many Type II patients subsequently reach a
point, however, when they require oral medications. As the disease progresses the oral agents often
fail. At that point, Type II patients join the Type I’s in having to take insulin injections.
In 1995 Europe and North America accounted for over 80% of the world insulin market because
the rates of diagnosis in those regions were high relative to other areas; because the incidence of
obesity was higher; and because a larger proportion of Type II patients tended to be treated with
insulin. Approximately 2% of the world population had diabetes, although many remained
undiagnosed with Type II diabetes. Of the diagnosed diabetic population, 10% were Type I (this
population was increasing by 2% to 3% a year) and 90% were Type II (increasing at 4% to 5% a year).
Early Discovery and Development of Insulin
Until 1921 there was no effective treatment for diabetes. Type I patients could expect to live
approximately one year from the time of diagnosis. Primary treatment was a starvation diet, based
upon the theory that less food generated less blood glucose and, therefore, prolonged life—slightly.
Diabetes wards in hospitals were populated by emaciated bodies, dying from a combination of
untreated diabetes and malnutrition. The physical appearance of patients in diabetes wards was later
likened to that of prisoners in the Nazi concentration camps of World War II.
1 Three symptoms typically induce a patient with diabetes to seek medical attention: 1) A rapid loss of weight, caused by their
inability to metabolize the food they eat (those with diabetes quite literally are starving at this stage, even though they eat
larger-than-normal amounts of food). 2) Constant thirst, and abnormally high frequency and amount of urination. This
happens because the body’s rate of urination is driven not only by how much water is in the body, but by how much glucose
the kidneys secrete. Hence, the body dehydrates. 3) Blurry vision, caused by the osmotic effects of high levels of glucose in the
eye on the lens of the eye.
2
This document is authorized for use only by di feng in SPRING 2019 (UNGER) THE INNOVATION PROCESS-1-1 taught by BARRY UNGER, Boston University from Jan 2019 to May 2019.
For the exclusive use of d. feng, 2019.
Eli Lilly and Company: Innovation in Diabetes Care
696-077
In 1921, four researchers from the University of Toronto—F.G. Banting, J.J.R. MacLeod, C.H. Best,
and J.B. Collip—had begun experimenting on dogs with pancreatic extractions of insulin. By 1922,
these researchers were injecting insulin extracted from animals into human patients with miraculous
results. Patients with diabetes could finally metabolize their food. Those who watched their nearly
instant recovery from starvation likened what they saw to the resurrection of the dead (Exhibit 3).
Banting and Best received the 1923 Nobel prize for this work.
The researchers from Toronto needed capital to begin consistent production in large quantities
and offered the Indianapolis-based drug maker, Eli Lilly and Company, an exclusive license to
produce and sell insulin in the United States. Lilly immediately began commercial development of
the product, and by the fall of 1923, 25,000 Americans were receiving insulin Lilly extracted from
pancreases of cows and pigs. Half of all Eli Lilly’s profits soon were derived from insulin sales. A
number of other companies subsequently began producing insulin in other regions of the world.
These included two Scandinavian companies, Nordisk and Novo, and the German chemical giant,
Hoechst. By 1995, Lilly and Novo-Nordisk (the two companies merged in 1989) dominated the world
market. Hoechst had a significant market position only in Germany.2 The sizes of the world insulin
market by region and the major competitors’ market shares, are detailed in Exhibit 4.
Subsequent Improvements in Insulin
Over the next 60 years, Lilly and its competitors improved their insulins along two dimensions.
The first was its purity. The second was in its “time profile”—matching the rate at which injected
insulin was absorbed into the blood3 with the rate at which glucose was absorbed into the
bloodstream.
Purity and the Development of Humulin
Parts per million (ppm) of proinsulin, the impurity which caused the majority of side effects from
insulin therapy, dropped from 50,000 ppm in 1921, to 10,000 ppm in 1970, and 10 ppm in 1980. Still,
though, all insulin products were derived from the pancreases of either cows or pigs (pork insulin
was closer in its molecular structure to human insulin than was beef-derived insulin). While similar
to that of humans, animal-derived insulin could never be molecularly equivalent to human insulin.
A fraction of a percent of the population with diabetes became resistant to insulin as a result.
In addition to this problem with animal insulins, Lilly feared an insulin shortage caused by a
combination of decreased red meat consumption and increased insulin usage. In response to these
concerns, Eli Lilly teamed with the biotechnology company Genentech to genetically engineer
bacteria that could synthesize and secrete human insulin. The result of this effort, branded in 1980 as
Humulin,® represented the supreme breakthrough in insulin product and process technology—a
100% pure insulin that was structurally identical to the insulin healthy humans produced. Lilly
invested over $700 million to build the first large-scale biotechnology plant in the world to produce
its Humulin.®4 The market responded poorly to Humulin®, though. Consumers resisted its
2 In 1995 there were also a few small producers of insulin in economically less developed countries as well. Because their
process technology for extracting insulin from animal pancreases was not sophisticated, many of their insulins were
notoriously impure.
3 Patients had to be careful to inject insulin into subcutaneous tissue (under the skin) and not directly into the bloodstream.
Doing so would suddenly lower blood sugar, precipitating unconsciousness, brain damage, and possibly death.
4 The pharmaceutical industry’s cost of developing a new drug, and of managing its progress through clinical trials to receive
regulatory approval, averaged over $300 million per approved drug in 1995.
3
This document is authorized for use only by di feng in SPRING 2019 (UNGER) THE INNOVATION PROCESS-1-1 taught by BARRY UNGER, Boston University from Jan 2019 to May 2019.
For the exclusive use of d. feng, 2019.
696-077
Eli Lilly and Company: Innovation in Diabetes Care
premium price and retailers were reluctant to add yet another set of SKUs to their already crowded
refrigerators of insulin products. The substitution of Humulin® for animal insulins occurred at a
slow pace, as a result, and it was not until the early 1990s that Humulin® accounted for 80% of Lilly’s
insulin volume. “In some developing countries the animal insulins they make have all sorts of
impurities in them,” noted Kathy Wishner, one of Lilly’s senior medical executives. “But all markets
are getting more sophisticated. Levels of impurities that were accepted a generation ago would not
be tolerated today. “Nonetheless,” added researcher Bruce Frank, “In retrospect the market was not
all that dissatisfied with highly purified pork insulin.”
The market’s sluggish response to Humulin®, its cost, and Lilly’s high share of the North
American insulin market (which often exceeded 85% and made it difficult to generate additional
volume through a new product such as Humulin®), all contributed to decreasing support for
continued investment in the insulin business at Lilly. Ron Chance, another Lilly researcher, recalled,
“At that point in time people were saying, ‘We can’t get any better than Humulin®. And we can’t
grow the business. It’s time to do something else.’ As a result, many of us went to other projects, like
the human growth hormone.”
Through the 1980s, particularly after Novo-Nordisk introduced its version of bio-engineered
human insulin in 1984, insulin became viewed essentially as a commodity product—the products of
Lilly, Novo and Hoechst were essentially identical in purity and efficacy. Nonetheless, because of the
high cost of clinical trials for new biotechnology products and the cost of an efficient manufacturing
facility, entry to the industry was limited, and as of 1995 the industry had not been afflicted by the
sorts of price battles that characterize the markets for many commodity products.
Action Profile
Glucose flows into the blood as food is digested. This flow rate, if graphed, resembles a bellshaped curve which reaches its peak flow rate about 1.5 hours after a meal. Some glucose is taken
into the liver, converted into a substance called glycogen, and stored there. To provide the body with
between-meal energy, the liver converts this stored glycogen back into glucose and secretes it at a
steady rate back into the bloodstream.
In nondiabetic persons, the body senses the amount of glucose flowing into the blood from these
two sources and signals the pancreas to secrete an offsetting amount of insulin to maintain about 100
milligrams of glucose per deciliter of blood. To achieve the same level of control over blood sugar
(and thereby avoid the complications of excessive or insufficient blood glucose), patients with
diabetes needed insulins that are absorbed from the subcutaneous injection site at two different
rates—one to be absorbed into the blood at a slow, constant rate, to offset the flow of glucose from the
liver, and the other to be absorbed into the blood at a faster, bell-shaped pace, to offset the flow of
glucose from digestion. Hence, many patients mixed a regular-acting and slow-acting insulin
together in their syringes for most injections. These rates of flow are depicted in Exhibit 5.
Because of these different flow patterns, patients who wanted to control their blood glucose levels
carefully had to take injections of regular insulin before every meal. Unfortunately, many found
injections to be inconvenient or uncomfortable, and took only one or two daily injections of longacting insulin.5 The resultant unhealthy pattern of insulin and glucose flows in the bloodstream of
these patients gave them high levels of blood glucose in the morning and low levels in the
afternoon—causing a high incidence of near-term emergencies and long-term complications.
5 Experts believed that only 20% of insulin-injecting patients actually took a shot before each meal.
4
This document is authorized for use only by di feng in SPRING 2019 (UNGER) THE INNOVATION PROCESS-1-1 taught by BARRY UNGER, Boston University from Jan 2019 to May 2019.
For the exclusive use of d. feng, 2019.
Eli Lilly and Company: Innovation in Diabetes Care
696-077
Unfortunately for the small proportion of patients taking pre-meal shots to keep better glucose
control, the fastest-acting available insulins followed a flow or action profile that was slightly slower
than the rate of glucose flow from digestion, as depicted in Exhibit 5.6 They therefore suffered a
temporary high level of blood glucose after meals unless they took their injection about 40 minutes
prior to eating. For most patients, however, taking an injection 40 minutes prior to each meal was
risky and inconvenient. For example, a diabetic could take an injection just prior to leaving work in
the evening, planning to drive home and start eating dinner an hour later. If caught in a long traffic
jam, however, the result could be disastrous. Similarly, if a diabetic took an advance injection prior to
eating out, but then was unable to eat the type or amount of food he or she had planned upon in the
injection, high or low blood glucose could result. Hence, many of even the most careful patients
simply injected themselves just before their meals, and lived with the mismatch in flow rates.
To respond to this problem, in the late 1980s Lilly launched an effort to develop an insulin that
could mimic more closely the normal physiologic secretion of insulin in people without diabetes. The
result, code-named Match, was successful. By 1994 it was clear that Match was absorbed into the
blood after injection at a rate that was much closer to the rate at which glucose was absorbed into the
blood after a meal. Consequently, patients using Match insulin in the clinical trials required for
regulatory approval were able to achieve better control of blood glucose levels after meals, reporting
fewer incidents of high and low blood glucose, compared to patients using regular insulin.
By the end of 1995, Match had been approved for general use by two governments. In addition to
launching Lilly’s new line of pens successfully, launching Match in each major market in 1996 was
high on Larry Ellingson’s innovation agenda. Novo-Nordisk’s own version of fast-acting insulin,
whose action profile reportedly was similar to Match, was two years behind Lilly in the regulatory
approval pipeline.
Exactly how Lilly should market Match was proving to be a vexing question for Ellingson’s team.
In the United States where Lilly’s share of the insulin market hovered near 80% in 1995, and it
appeared that every vial of Match that Lilly sold to an existing customer would directly supplant a
vial of regular insulin which that customer had been purchasing. In Europe and Japan, where Lilly’s
and Novo’s market share positions essentially were reversed, there was greater immediate-term
potential for Lilly to win new customers with Match. Remembering Lilly’s struggles to achieve
premium pricing for Humulin®, Ellingson was also concerned that Match be capable of sustaining a
premium price relative to regular insulin, given its better performance.
In addition to its fast-acting Match insulin, Lilly researchers had been working on a new, longacting “basal?…
Purchase answer to see full
attachment