Item 8. Financial Statements and Supplementary Data

UNITED STATES
SECURITIES AND EXCHANGE COMMISSION
Washington, D.C. 20549

Form 10-K

Commission file number 0-17999

ImmunoGen, Inc.

         Securities registered pursuant to Section 12(b) of the Act:

         Indicate by check mark if the registrant is a well-known seasoned issuer, as defined in Rule 405 of the Securities Act. o Yes     ý No

         Indicate by check mark if the registrant is not required to file reports pursuant to Section 13 or Section 15(d) of the Act. o Yes    ý No

         Indicate by check mark whether the registrant (1) has filed all reports required to be filed by Section 13 or 15(d) of the Securities Exchange Act of 1934 during the preceding 12 months (or for such shorter period that the registrant was required to file such reports), and (2) has been subject to such filing requirements for the past 90 days. ý Yes    o No

         Indicate by check mark if disclosure of delinquent filers pursuant to Item 405 of Regulation S-K (§229.405 of this chapter) is not contained herein, and will not be contained, to the best of registrant's knowledge, in definitive proxy or information statements incorporated by reference in Part III of this Form 10-K or any amendment to this Form 10-K. ý

         Indicate by check mark whether the registrant is a large accelerated filer, an accelerated filer, a non-accelerated filer or a smaller reporting company. See definitions of "large accelerated filer," "accelerated filer," and "smaller reporting company" in Rule 12b-2 of the Exchange Act. (Check one):

         Indicate by check mark whether the registrant is a shell company (as defined in Rule 12b-2 of the Exchange Act). o Yes    ý No

         Aggregate market value, based upon the closing sale price of the shares as reported by the NASDAQ Global Market, of voting stock held by non-affiliates at December 31, 2007 $181,598,190 (excludes shares held by executive officers, directors, and beneficial owners of more than 10% of the Company's common stock). Exclusion of shares held by any person should not be construed to indicate that such person possesses the power, direct or indirect, to direct or cause the direction of management or policies of the registrant, or that such person is controlled by or under common control with the registrant. Common Stock outstanding at August 28, 2008: 50,785,760 shares.

Table of Contents

ImmunoGen, Inc.
Form 10-K
TABLE OF CONTENTS

2

Table of Contents

Item 1.    Business

        In this Annual Report on Form 10-K, ImmunoGen, Inc. (ImmunoGen, Inc., together with its subsidiaries, is referred to in this document as "we", "us", "ImmunoGen", or the "Company"), incorporates by reference certain information from parts of other documents filed with the Securities and Exchange Commission. The Securities and Exchange Commission allows us to disclose important information by referring to it in that manner. Please refer to all such information when reading this Annual Report on Form 10-K. All information is as of June 30, 2008 unless otherwise indicated. For a description of the risk factors affecting or applicable to our business, see "Risk Factors," below.

The Company

        We develop novel, targeted therapeutics for the treatment of cancer using our expertise in cancer biology, monoclonal antibodies, and small molecule cytotoxic, or cell-killing, agents. Our Tumor-Activated Prodrug, or TAP, technology uses antibodies to deliver a potent cytotoxic agent specifically to cancer cells, and consists of a monoclonal antibody that binds specifically to a cancer target with one of our proprietary cell-killing agents attached. The antibody component enables a TAP compound to bind specifically to cancer cells that express a particular target antigen and the cytotoxic agent serves to kill the cancer cell. Our TAP technology is designed to enable the creation of highly effective, well-tolerated anticancer products.

        We believe that our TAP technology along with our expertise in antibodies has made us a leader in the field of "armed antibody" therapeutics for the treatment of cancer. We achieved this position through the creation of our own anticancer compounds and through out-licenses of our TAP technology to other companies. The out-licensing of our TAP technology allows us to expand the number of anticancer therapeutics in which we have a financial interest by enabling the creation of TAP compounds with antibodies proprietary to other companies to which we do not have access for our own development programs. There are now multiple anticancer compounds in clinical trials utilizing our TAP and/or antibody technology. Our out-licensing partners include: sanofi-aventis, Genentech, Inc., Amgen Inc., Biogen Idec Inc. and Biotest AG.

        We believe that the key initiatives central to our future success are:

•

Develop our own proprietary products.  We currently have three TAP compounds in clinical testing: IMGN242, a potential treatment for stomach cancer and other CanAg-expressing malignancies; IMGN901, a potential treatment for multiple myeloma, small-cell lung cancer or SCLC, ovarian cancer and other CD56-expressing cancers; and IMGN388, a potential treatment for solid tumors including melanomas, sarcomas and many carcinomas. We are advancing these compounds and also are using our cancer biology and antibody expertise, along with our TAP technology, to develop additional proprietary compounds. Several compounds are in the research assessment phase from which we will determine their suitability to advance to preclinical development over the next twelve months.



•

Support and expand our collaborative arrangements.  Part of our business model is to out-license our TAP technology to other companies to enable its use with antibodies to targets proprietary to these companies to which we do not have access for our own product programs. These licenses provide us with upfront payments and the opportunity to earn milestone payments, research and manufacturing revenue, and royalties on the sales of any resulting products. For example, Genentech created T-DM1, now in Phase II clinical testing, through one of these collaborative agreements. We intend to continue to out-license our TAP technology.

We also establish other types of collaborative arrangements to expand the opportunity for us to earn a return from our product programs, our technology, and our capabilities. We have licensed sanofi- aventis expanded access to our antibody humanization technology, which was developed to

3

Table of Contents

enable monoclonal antibodies initially of murine origin to avoid detection by the human immune system. We have entered into arrangements with two of our collaborators—Genentech and sanofi-aventis—to assist them in scaling up their manufacturing processes for TAP compounds. In 2003, we entered into a broad collaboration with sanofi-aventis that has been providing us with committed research funding and has enabled us to work together with sanofi-aventis to develop additional TAP and naked antibody anticancer compounds. Three compounds—AVE9633, AVE1642 and SAR3419—have entered clinical testing through this collaboration, and we expect additional compounds to follow over the next several years.


•

Support our TAP technology to maintain our strong position in our field.  We have developed highly potent cell-killing agents designed specifically for attachment to antibodies for selective delivery to cancer cells, and have a portfolio of linkers to affix our cytotoxic agents to antibodies. These cell-killing agents and linkers provide us and our collaborators the flexibility to select the design that works best for each antibody and target. More antibody-cytotoxic agent compounds have advanced into clinical testing using our technology than that of any other company. We continue to invest in our TAP and antibody technology and to conduct research to develop additional cell-killing agents and linkers to retain our strong position in the field.

        We were organized as a Massachusetts corporation in March 1981. Our principal offices are located at 830 Winter Street, Waltham, Massachusetts (MA) 02451, and our telephone number is (781) 895-0600. We maintain a website at www.immunogen.com, where certain information about us is available. Please note that information contained on the website is not a part of this document. Our Annual Reports on Form 10-K, Quarterly Reports on Form 10-Q, Current Reports on Form 8-K, and any amendments to those reports are available free of charge through the "Investor Information" section of our website as soon as reasonably practicable after those materials have been electronically filed with, or furnished to, the Securities and Exchange Commission. We have adopted a Code of Corporate Conduct that applies to all our directors, officers and employees and a Senior Officer and Financial Personnel Code of Ethics that applies to our senior officers and financial personnel. Our Code of Corporate Conduct and Senior Officer and Financial Personnel Code of Ethics are available free of charge through the "Investor Information" section of our website.

Our TAP Technology

        Traditional chemotherapeutic agents typically kill any rapidly-dividing cell, including healthy cells. This can limit the ability of these agents to be dosed to full therapeutic potential, and can result in significant adverse side effects. Monoclonal antibodies, in contrast, can be made that bind specifically to targets that can be found predominantly or exclusively on cancer cells, thereby allowing the antibody to attach to these cancer cells. Many antibodies that bind to cancer cells, however, have been found to have little or no therapeutic effect.

        Our TAP technology uses antibodies to deliver one of our highly potent cell-killing agents specifically to cancer targets. Our TAP technology can be used with antibodies that have anticancer activity of their own to create compounds with enhanced anticancer activity. In addition, our TAP technology can be used with antibodies that lack anticancer activity to achieve an effective therapeutic, since the attached cell-killing agent can kill the cancer cell. Therefore, we believe our TAP technology can be used to create effective, well-tolerated anticancer therapeutics with antibodies that may, as well as with those that may not, have the potential to become commercial products as naked antibodies.

        We developed our cell-killing agents specifically for antibody-directed delivery to cancer targets. Our cell-killing agents are:

•

Potent.  Our cytotoxic agents are 1,000- to 10,000-fold more potent than traditional chemotherapeutic agents, and are thus capable of killing cancer cells when the agents are present at low concentrations. This is important for an agent delivered to a cancer cell attached

4

Table of Contents

to an antibody, as generally only a small amount of antibody, and therefore, the attached cell-killing agent, will reach the cancer cells. The agents used in the TAP compounds currently in clinical or preclinical development are our proprietary derivatives of maytansine, a highly potent molecule that interferes with the activity of tubulin, a substance necessary for cells, including cancer cells, to successfully divide.



•

Attachable.  Our cytotoxic agents can be attached to an antibody using one of our proprietary "linkers." Our linkers are designed to achieve a bond between the agent and the antibody that remains intact while the TAP compound is circulating in the bloodstream, rendering the cytotoxic agent inactive, but then enables the cytotoxic agent to exhibit its full potency once inside a cancer cell.



•

Non-immunogenic.  Our cytotoxic agents are small molecules rather than protein-based toxins to avoid the stimulation of an immune response that would limit the activity of TAP compounds upon repeat administration. To date, our TAP compounds have been tested in over 400 patients with no evidence of an immunogenic response in any patient.



•

Producible.  Our cytotoxic agents can be readily manufactured and our supplier produces these agents for us in a manner that can be scaled-up to commercial production quantities. We have extensive experience attaching our agents to antibodies and have produced numerous TAP compounds at our manufacturing facility in Norwood, MA for use in clinical testing.



•

Protectable.  We patent our cytotoxic agents and related derivatives to protect these assets. We hold U.S. patents claiming a process for the preparation of certain cell-killing agents, including our maytansinoid agents, and claiming methods of preparation of conjugates composed of our cell-killing and cell-binding agents.

        We have developed alternative maytansinoid cell-killing agents (such as DM1 and DM4) and linkers. This enables us to create highly-hindered disulfide bonds, less-hindered disulfide bonds and also a non-reducible or "non-cleavable" thioether bond. This provides us and our collaborators with flexibility in the construction of TAP compounds as the best design for each TAP compound varies depending upon the particular antibody and its target. We have active programs to further expand our portfolio of linkers and cell-killing agents, and recently unveiled a new family of ImmunoGen linkers that provide enhanced activity against multi-drug resistant cancers. In recent years, we have gained increasing recognition by our collaborative partners for the depth of our expertise in the design, evaluation and development of antibody-cytotoxic agent compounds.

        Additionally, we have established capabilities and expertise with monoclonal antibodies. We have extensive experience in cancer biology and in the evaluation of potential targets for antibody-based anticancer treatments. We can create monoclonal antibodies for promising targets, and using our patented humanization technology we can modify these antibodies so that the human immune system is unable to detect them. We also have considerable expertise in functions critical to the advancement of an antibody-based product from the laboratory to the clinic, including cell-line development, preclinical evaluation and process development.

Product Candidates

        The following table summarizes the antigen target, cancer(s) expressing the target, and development stage for compounds in development by us and our collaborators. The results from preclinical testing and early clinical trials may not be predictive of results obtained in subsequent clinical trials and there can be no assurance that our or our collaborators' clinical trials will

5

Table of Contents

demonstrate the level of safety and efficacy of any product candidates necessary to obtain regulatory approval.

⁽¹⁾

Types of cancers that express the target antigen. Not all tumors of any given type may express the antigen target.

⁽²⁾

Compounds in clinical testing are being assessed in patients whose cancer expresses the target antigen. Compounds that are not in clinical testing and have an undisclosed status are listed as research/preclinical.

⁽³⁾

Sanofi-aventis has informed us that they plan to discontinue the development of AVE9633 after the completion of the treatment of study patients.

⁽⁴⁾

Naked antibody.

⁽⁵⁾

Patient recruitment underway.

Trastuzumab-DM1 (T-DM1)

        T-DM1 is in development by Genentech for the treatment of HER2 postive metastatic breast cancer. It comprises Genentech's anti-HER2 antibody, trastuzumab, with our DM1 attached, and was developed under a HER2-specific 2000 license agreement between the companies. Trastuzumab is the active, antibody component of the anticancer compound marketed as Herceptin®.

        In January 2006, Genentech informed us that the Investigational New Drug, or IND, application for T-DM1 had become effective, triggering a $2 million milestone payment to us. T-DM1 began Phase I evaluation in April 2006. Genentech began Phase II evaluation of T-DM1 in July 2007 and we earned a $5 million milestone payment with this event.

6

Table of Contents

        Beginning in December 2006, Genentech began reporting data from the Phase I evaluation of T-DM1, which was conducted in patients who all had HER2-positive metastatic breast cancer that progressed on treatment with Herceptin plus chemotherapy. The most comprehensive presentation of these findings was at the annual meeting of the American Society of Clinical Oncology, or ASCO, in June 2008. In Phase I testing, 12 of the 15 patients treated with T-DM1 given every three weeks at its maximum tolerated dose, or MTD, had either stable disease or an objective response by RECIST criteria. Median progression-free survival in these patients was 9.8 months. At its MTD, T-DM1 was found to be generally well tolerated. It was found that dosing T-DM1 weekly yielded comparable efficacy and safety data as administration once every three weeks. It should be noted that findings in the limited number of patients in these studies may not be predictive of the findings in a larger population.

        In late calendar 2007, Roche exercised their right with Genentech to opt-in on the development and commercialization of T-DM1.

        In July 2008, Genentech disclosed in their quarterly conference call that patient enrollment had been completed in the Phase II study begun in July 2007, and that they expect to report interim findings from this study at the ASCO Breast Cancer Symposium that is being held in September 2008. In this call, Genentech also discussed their development plans for T-DM1 for the treatment of HER2-positive metastatic breast cancer:

•

Genentech reported that they plan to initiate a Phase II trial assessing T-DM1 as a third-line treatment for this cancer during the second half of 2008 and that if the results from this study are compelling, that they will discuss an earlier approval pathway with the US Food and Drug Administration, or FDA;



•

Genentech discussed that they plan to initiate a Phase II trial assessing T-DM1 as a first-line treatment for this cancer during the second half of 2008, comparing T-DM1 used alone with Herceptin used together with Taxotere (docetaxel);



•

Genentech also noted that they plan to make a Phase III decision in 2008 related to potentially also evaluating T-DM1 as a second-line treatment for this cancer.

IMGN901

        Our IMGN901 TAP compound, previously called huN901-DM1, targets the antigen known as CD56. CD56-expressing cancers include many cases of multiple myeloma as well as other hematological malignancies, small-cell lung cancer, or SCLC, ovarian cancer and other cancers of neuroendocrine origin. IMGN901 consists of our CD56-binding antibody huN901 with our DM1 cell-killing agent attached.

        We currently have three clinical trials underway with IMGN901. Study 001 is evaluating the compound for the treatment of SCLC. Study 002 is evaluating it for the treatment of CD56-expressing solid tumors, including SCLC. Study 003 is evaluating IMGN901 for the treatment of CD56-expressing multiple myeloma. IMGN901 is administered alone, as monotherapy, in all three of these trials as part of establishing the safety of the agent. We expect to report clinical findings in both multiple myeloma and solid tumors in the fourth quarter of 2008.

        We believe development of IMGN901 for the treatment of multiple myeloma, of which approximately 70% of such cases express CD56, represents a faster pathway to marketing approval for this compound than its development for the treatment of SCLC or other solid tumors. We also believe that we can develop IMGN901 faster for use in combination with an approved multiple myeloma agent than as monotherapy in light of current trends in the management of multiple myeloma. Thus, after we establish the MTD of IMGN901 as monotherapy in multiple myeloma patients, we intend to initiate a

7

Table of Contents

Phase I/II trial evaluating it used in combination with an approved agent in a less treatment-resistant patient population. We expect this combination trial to begin in the first half of 2009.

IMGN242

        Our TAP product candidate IMGN242, previously called huC242-DM4, consists of our CanAg-binding antibody, huC242, with our DM4 cell-killing agent attached. The CanAg antigen is found on many gastrointestinal tumors including gastric (stomach), pancreatic and colorectal cancers, as well as many non-small-cell lung cancers.

        Our IMGN242 Phase I clinical trial is designed to assess the safety and tolerability of the compound and to establish its MTD when administered once every three weeks to patients with CanAg-expressing cancers. To qualify for enrollment, patients must have failed the approved treatments for their particular cancer. Prior to the establishment of the MTD, patients with any level of CanAg expression were eligible for enrollment in the trial. Once the MTD was established, enrollment was limited to patients with cancer that strongly and consistently expresses CanAg.

        In July 2007, we initiated a Phase II clinical trial assessing IMGN242 for the treatment of CanAg-expressing gastric cancer. The patients in this trial have metastatic or locally-advanced CanAg-expressing gastric or gastroesophageal cancer that has failed to respond to front-line therapy. We estimate that approximately 50% of such cancers express CanAg. We reported IMGN242 pharmacokinetic/pharmacodynamic data at the ASCO annual meeting in June 2008. The poster presented described one of the first patients treated with IMGN242 in this study who had an encouraging response to treatment, but did not qualify as an objective response by RECIST criteria. In early 2008, the protocol for this study was amended to reduce the dose of IMGN242 administered in patients with low plasma levels of CanAg. The current design of the study is to evaluate IMGN242 in 23 patients, treated at 126 mg/m², with CanAg-expressing gastric or gastroesophageal cancer. If an objective response by RECIST criteria is seen in any of these 23 patients, the study is to be expanded to approximately 40 patients. Otherwise, the study will end. Our goal is to complete enrollment of the 23 patients by the end of our 2009 fiscal year on June 30, 2009.

IMGN388

        Our TAP compound IMGN388 comprises our DM4 cell-killing agent attached to an integrin-targeting antibody developed by Centocor. The target for this compound is found on the cancer cells of melanomas, sarcomas and many carcinomas, including lung, bladder, renal cell and thyroid carcinomas. It is also found on endothelial cells that are engaged in forming new blood vessels. All solid tumors require the formation of new blood vessels in order to grow. Thus, IMGN388 can potentially attack tumors in two ways: (1) by attaching to cancerous cells that express its target and killing them; and (2) by disrupting the formation of the new blood vessels that a solid tumor needs to grow. ImmunoGen initiated Phase I testing of IMGN388 in July 2008 in patients with solid tumors. Centocor has opt-in rights for this compound.

AVE9633

        This TAP compound was created by us and licensed to sanofi-aventis from our preclinical pipeline as part of a broader collaboration. It comprises our huMy9-6 antibody, which targets a CD33-binding antibody that was developed and humanized by us and our DM4 cell-killing agent. In March 2005, sanofi-aventis informed us that patient dosing with AVE9633 had begun in a Phase I clinical trial, triggering a $2 million milestone payment to us. This compound currently is being evaluated in its third Phase I clinical study for the treatment of acute myeloid leukemia. Sanofi-aventis has informed us that they plan to discontinue the development of AVE9633 after the completion of the treatment of study patients.

8

Table of Contents

AVE1642

        We created this naked-antibody compound and licensed it to sanofi-aventis from our preclinical pipeline as part of our broader collaboration with sanofi-aventis. AVE1642 binds to the IGF-1 receptor. Its purpose is to block cancer cells from using an IGF-1-mediated survival pathway to withstand exposure to chemotherapy treatments. Thus, AVE1642 is intended to be used in combination with chemotherapeutic agents to kill cancer cells. It has potential utility for the treatment of certain solid tumors and hematological malignancies. In October 2006, sanofi-aventis informed us that patient dosing had begun in the Phase I evaluation of AVE1642, triggering a $2 million milestone payment to us. Phase I studies are underway to conduct an initial assessment of the compound when used in combination with approved chemotherapeutic agents. Phase I data were reported at the ASCO annual meeting in June 2008 that showed AVE1642 demonstrates promising tolerability and activity used in combination with docetaxel for the treatment of solid tumors.

SAR3419

        Sanofi-aventis also licensed rights to this TAP compound as part of our broader collaboration. SAR3419 consists of a CD19-targeting antibody developed and humanized by us and our DM4 cell-killing agent. CD19 is associated with certain B-cell hematological malignancies, including non-Hodgkin's lymphoma. Sanofi-aventis initiated a dose-escalation Phase I clinical trial of SAR3419 in October 2007, triggering a $1 million milestone payment to us. The trial is designed to assess the safety and tolerability of SAR3419 and to establish its MTD when administered once every three weeks to patients with relapsed or refractory B-cell non-Hodgkin's lymphoma. Preliminary evidence of anti-lymphoma activity is also being assessed. The first findings from this study are expected to be reported at a medical conference in December 2008.

BIIB015

        This TAP compound was created by Biogen Idec under a 2004 license that grants Biogen Idec the exclusive right to use our maytansinoid TAP technology with antibodies that target Cripto. BIIB015 consists of Biogen Idec's Cripto-binding antibody and our DM4 cell-killing agent. Biogen Idec submitted the IND for this compound to the FDA in February 2008, triggering a $1.5 million milestone payment to us. BIIB015 advanced into Phase I testing in the summer of 2008.

BT-062

        This TAP compound was created by Biotest under the 2006 license that grants Biotest the exclusive right to use our maytansinoid TAP technology with antibodies to an undisclosed target found on multiple myeloma and certain other cancers. Biotest submitted the IND for this compound to the FDA in March 2008 and patient recruitment is underway for its open Phase I study. We have opt-in rights on BT-062 in the U.S.

SAR566658

        This TAP compound is in development through our collaboration with sanofi-aventis. It consists of an antibody that binds to a target found on breast and ovarian cancers as well as on other solid tumors and our DM4 cell-killing agent.

SAR650984

        SAR650984 is a naked antibody compound developed in our collaboration with sanofi-aventis. Its target, CD38, is expressed on a number of hematological malignancies.

9

Table of Contents

Other Compounds in Development

        Additional product candidates using our technology are in various stages of preclinical research and development internally and at our collaborators.

Incidence of Relevant Cancers

        Cancer remains a leading cause of death worldwide, and is the second leading cause of death in the U.S. The American Cancer Society projects that 1.4 million new cases of cancer will be diagnosed in the U.S. in 2008 and that 566,000 people will die from various cancers in the U.S. in 2008. The total number of people living with cancer significantly exceeds the number of patients diagnosed with cancer in a given year as patients can live with cancer for a year or longer. Additionally, the potential market for anticancer drugs exceeds the number of patients treated as many types of cancer typically are treated with multiple compounds at the same time. Additionally, patients often receive multiple drug regimens sequentially, either to treat or help prevent recurrence of the disease.

        In July 2007, we began Phase II evaluation of our TAP product candidate, IMGN242, for the treatment of CanAg-expressing gastric cancer. We estimate that approximately half of all gastric cancer tumors express CanAg. Globally, gastric cancer is one of the leading causes of death in both high- and middle-income countries, according to the World Health Organization. It is particularly common among Asian populations, but occurs across ethnicities. The American Cancer Society estimates that in the U.S., in 2008 alone, 21,500 new cases of gastric cancer will be diagnosed and 10,880 people will die from the disease.

        We are assessing our IMGN901 compound for the treatment of multiple myeloma, SCLC, and other CD56-expressing solid tumors. Our highest priority is the development of IMGN901 for the treatment of multiple myeloma. According to the American Cancer Society, approximately 20,000 new cases of multiple myeloma will be diagnosed in the U.S. in 2008, and close to 11,000 people will die from the disease in 2008. Based on research conducted, we believe that approximately 70% of multiple myeloma cases express the CD56 antigen targeted by IMGN901.

        We are assessing IMGN388 for the treatment of solid tumors. IMGN388 may be able to be used for solid tumors that do not express its target antigen as well as ones that do because of the potential ability of the compound to interfere with the formation of the new blood vessels that all solid tumors require in order to grow. Over 1 million of the new cancer cases diagnosed in a year are solid tumors.

        In recent years, several antibody-based anticancer drugs—such as Herceptin®, Rituxan®, Avastin®, and Erbitux®—have enjoyed considerable commercial success, as have other targeted anticancer agents.

Outlicenses and Collaborations

        As part of our business strategy to expand the use and financial return from our TAP technology, we enter into license agreements with third parties where we grant them the exclusive right to use our TAP technology with their antibodies to proprietary or non-proprietary targets. In some cases, we have out-licensed rights to our own TAP compounds to companies with product development and commercialization capabilities that we desired to access. In exchange, we are entitled to receive upfront fees, potential milestone payments and royalties on any product sales. Our principal out-licenses and collaborative agreements are described below.

sanofi-aventis

        In July 2003, we entered into a broad collaboration agreement with sanofi-aventis to discover, develop and commercialize antibody-based anticancer therapeutics.

10

Table of Contents

        The agreement provides sanofi-aventis with worldwide commercialization rights to new anticancer therapeutics developed to targets included in the collaboration, including the right to use our TAP technology and our humanization technology in the creation of therapeutics to these targets. The product candidates (targets) currently in the collaboration include AVE9633 (CD33), AVE1642 (IGF-1R), SAR3419 (CD19), SAR566658 (DS6), SAR650984 (CD38) and additional compounds at earlier stages of development that have yet to be disclosed.

        The collaboration agreement entitles us to receive milestone payments potentially totaling $21.5 million to $30.0 million, per antigen target, for each therapeutic developed under the collaboration agreement. To date we have earned a $2 million milestone payment in March 2005 with the start of clinical testing of AVE9633, a $2 million milestone payment in October 2006 with the start of clinical testing of AVE1642, a $500,000 milestone payment in September 2004 for a preclinical milestone related to SAR3419, a $1 million milestone payment in October 2007 with the start of clinical testing of SAR3419, a $500,000 milestone payment in December 2007 for a preclinical milestone related to SAR650984 and a $500,000 milestone payment in March 2008 for a preclinical milestone related to SAR566658.

        The agreement also entitles us to royalties on the commercial sales of any resulting products, if and when such sales commence. Sanofi-aventis is responsible for the cost of the development, manufacturing and marketing of any products created through the collaboration. We are reimbursed for any preclinical and clinical materials that we make under the agreement. The collaboration agreement also provides us an option to certain co-promotion rights in the U.S. on a product-by-product basis. The terms of the collaboration agreement allow sanofi-aventis to terminate our co-promotion rights if there is a change of control of our company.

        The overall term of the agreement extends to the later of the latest patent to expire or twelve years after the latest launch of any product discovered, developed and/or commercialized under the agreement. Sanofi-aventis paid us an upfront fee of $12.0 million in August 2003. Inclusive of its extensions, the agreement entitled us to receive committed research funding totaling $79.3 million over the five years of the research collaboration. The 2003 agreement committed sanofi-aventis to a minimum of $50.7 million of committed research funding during the three-year research period. Under the 2003 agreement, sanofi-aventis was granted the option, whereby upon giving 12 months' advance notice for each, they could request that we extend the research program for two additional 12-month periods. In August 2005, sanofi-aventis exercised their contractual right to extend the term of their research program with us and committed to fund $18.2 million in additional research and support over the 12-month period following September 1, 2006. In August 2006, sanofi-aventis exercised its remaining option to extend the term of the research collaboration with us for another year, and committed to pay us a minimum of $10.4 million in additional research support over the twelve months beginning September 1, 2007.

        In October 2006, sanofi-aventis licensed non-exclusive rights to use our proprietary resurfacing technology to humanize antibodies to targets not included in the collaboration, including antibodies for non-cancer applications. This license provides sanofi-aventis with the non-exclusive right to use our proprietary humanization technology through August 31, 2011 with the right to extend for one or more additional periods of three years each by providing us with written notice prior to expiration of the then-current license term. Under the terms of the license, we are entitled to a $1 million license fee, half of which was paid upon contract signing and the second half was paid in August 2008, and in addition, we are entitled to receive milestone payments potentially totaling $4.5 million for each antibody humanized under this agreement and also royalties on commercial sales, if any.

        In August 2008, sanofi-aventis exercised its option under a 2006 agreement for expanded access to our TAP technology. The exercise of this option enables sanofi-aventis to evaluate, with certain restrictions, our maytansinoid TAP technology with antibodies to targets not included in the existing

11

Table of Contents

research collaboration between the companies and to license the exclusive right to use the technology to develop products to specific targets based on the terms in the 2006 agreement. We are entitled to earn upfront and milestone payments potentially totaling $32 million per target for each compound developed under the 2006 agreement, as well as royalties on commercial sales. We are also entitled to manufacturing payments for any materials made on behalf of sanofi-aventis. We received $3.5 million with the exercise of this option in August 2008, in addition to the $500,000 we received in December 2006 with the signing of the option agreement. The agreement has a three-year term from the date of the exercise of the option and can be renewed by sanofi-aventis for one additional three-year term by payment of a $2 million fee.

Genentech, Inc.

        In May 2000, we entered into two separate agreements with Genentech. The first agreement grants Genentech an exclusive license to our maytansinoid TAP technology for use with antibodies, such as Herceptin (trastuzumab), that target HER2. Under the terms of this agreement, Genentech has exclusive worldwide rights to develop and commercialize maytansinoid TAP compounds with antibodies that target HER2. Genentech is responsible for the manufacturing, product development and marketing of any products resulting from the agreement. We are reimbursed for any preclinical and clinical materials that we manufacture under the agreement. We received a $2 million non-refundable payment upon execution of the agreement. In addition to royalties on net sales if and when they occur, the terms of the agreement include other payments based upon Genentech's achievement of milestones. In May 2006, we and Genentech amended this agreement. This amendment increases the potential milestone payments to us under this agreement and the potential royalties to us on any HER2 targeting TAP compound that may be developed by Genentech, including T-DM1. Assuming all benchmarks are met under this agreement, we will receive up to $44 million in milestone payments. In January 2006, Genentech notified us that the IND application for T-DM1 submitted to the FDA had become effective. Under the terms of this agreement, this event triggered a $2 million milestone payment to us. Genentech began Phase II evaluation of T-DM1 in July 2007 and we earned a $5 million milestone payment with this event.

        In May 2000 we entered into a second agreement with Genentech. This second agreement provided Genentech with the right to test our maytansinoid TAP technology with Genentech antibodies to a defined number of targets on an exclusive basis for a specified period of time, known as the "option period," and to take exclusive licenses for individual targets on agreed upon terms to use our maytansinoid TAP technology to develop products. We received a non-refundable technology access fee of $3 million when we entered into this five-year agreement in May 2000, and an additional technology access fee of $2 million when Genentech renewed this agreement in April 2005 for the one additional three-year period allowed. Genentech no longer has the right to designate new targets under this "right to test" agreement, although there are option periods with respect to previously-designated targets that remain in effect for the remainder of the respective option periods.

        Under this agreement, in April 2005, July 2005 and December 2005, Genentech licensed exclusive rights to use our maytansinoid TAP technology with antibodies that target three undisclosed targets. Under the terms defined in the 2000 "right-to-test" agreement, for each license we received a $1 million license fee and may receive up to $38 million in milestone payments. We are also entitled to receive royalties on the sales of any resulting products. Genentech is responsible for the development, manufacturing, and marketing of any products resulting from these licenses.

Biogen Idec Inc.

        In October 2004, we entered into a development and license agreement with Biogen Idec MA Inc., or Biogen. Under the terms of the agreement, Biogen received exclusive worldwide rights to develop and commercialize anticancer therapeutics using antibodies to the tumor cell target, Cripto, and a

12

Table of Contents

maytansinoid cell-killing agent developed by us. Biogen is responsible for the research, development, manufacturing, and marketing of any products resulting from the license. Under the terms of the agreement, we received from Biogen an upfront payment of $1 million upon execution of the agreement. In January 2008, Biogen submitted an IND for their TAP compound, BIIB015, to the FDA. We earned a $1.5 million milestone payment with this event. Biogen began Phase I evaluation of BIIB015 in June 2008. Assuming all benchmarks are met, we could receive up to $42 million in milestone payments under this agreement. We are also entitled to receive royalties on net sales of resulting products. We will also receive compensation from Biogen for product development research done on its behalf, as well as for the production of preclinical and clinical materials.

Biotest AG

        In July 2006, we entered into a development and license agreement with Biotest AG, or Biotest. The agreement grants Biotest exclusive rights to use our TAP technology with antibodies that target a specific antigen that occurs on multiple myeloma cells to create anticancer therapeutics. In March 2008, Biotest submitted an IND to the FDA for a TAP compound, BT-062, developed under this agreement to the FDA. Under the agreement, we received a $1 million upfront payment upon execution of the agreement, and could potentially receive up to $35.5 million in milestone payments, and royalties on the sales of any resulting products. We receive payments for manufacturing any preclinical and clinical materials made at the request of Biotest. The agreement also provides us with the right to elect to participate, at specific stages during the clinical evaluation of any compound created under this agreement, in the U.S. development and commercialization of that compound in lieu of receiving royalties on U.S. sales of that product and the milestone payments not yet earned. We can exercise this right by making a payment to Biotest of an agreed-upon fee of $5 million or $15 million, depending on the stage of development. Upon exercise of this right, we would share equally with Biotest the associated costs of product development and commercialization in the U.S. along with the profit, if any, from U.S. product sales.

Amgen, Inc.

        In September 2000, we entered into a ten-year collaboration agreement with Amgen, Inc., or Amgen. The agreement provides Amgen with the right to test our maytansinoid TAP technology with antibodies to a defined number of targets on an exclusive and non-exclusive basis for a specified period of time, known as the "option period," and to take exclusive licenses for individual targets on agreed upon terms to use our maytansinoid TAP technology to develop products. We received a $5 million technology access fee in September 2000 and are entitled to potential milestone payments and royalties on net sales of products developed under any licenses taken under this agreement, if and when such sales commence. In addition, on September 7, 2000, Amgen purchased $15 million of our common stock in accordance with the agreement. We understand that these shares were sold in fiscal 2006. In April 2007 and July 2008, we granted Amgen a non-exclusive option and exclusive option, respectively, to test our TAP technology with antibodies to specific targets. For each option taken, Amgen paid us a nominal fee. Under this agreement, there can be option periods in effect that extend beyond the expiration of this right to test agreement in September 2010.

In-Licenses

        From time to time we may in-license certain rights to targets or technologies, in conjunction with our internal efforts to develop both TAP and naked-antibody products and related technologies. In exchange, we may be obligated to pay upfront fees, potential milestone payments and royalties on any product sales.

13

Table of Contents

Centocor, Inc.

        In December 2004, we entered into a development and license agreement with Centocor, Inc., or Centocor, a wholly-owned subsidiary of Johnson & Johnson. Under the terms of this agreement, Centocor was granted exclusive worldwide rights to develop and commercialize anticancer therapeutics that comprise an antibody developed by Centocor that binds to an integrin cancer target and a maytansinoid cell-killing agent developed by us. Under the terms of the agreement, we received a non-refundable upfront payment of $1 million upon execution of the agreement.

        In December 2007, we licensed from Centocor the exclusive, worldwide right to develop and commercialize a TAP compound, IMGN388, that consists of an integrin-binding antibody developed by them and one of our maytansinoid cell-killing agents. This license reallocates the parties' respective responsibilities and financial obligations from the license referenced above. Centocor has the right to opt-in on future development and commercialization of IMGN388 at an agreed-upon stage in early clinical testing. Should Centocor not exercise this right, Centocor would be entitled to receive milestone payments potentially totaling $30 million, with the first payment due upon the completion of a successful Phase III trial, and also royalties on IMGN388 sales, if any. In this event, ImmunoGen has the right to obtain a new partner for IMGN388, with certain restrictions. Should Centocor exercise its opt-in right, ImmunoGen would receive an opt-in fee and be released from its obligation to pay Centocor any milestone payments or royalties on sales. Both companies would contribute to the costs of developing the compound. The two companies would share equally any profits on the sales of the compound in the U.S. and ImmunoGen would receive royalties on any international sales. The companies have agreed to share certain third-party expenses. In June 2008, the FDA approved the IND application for IMGN388. This event triggered a $1 million milestone payment to a third party, half of which is to be paid by ImmunoGen.

Other Licenses

        We also have licenses with third parties, including other companies and academic institutions, to gain access to techniques and materials for drug discovery and product development and the rights to use those techniques and materials to make our product candidates. These licenses include rights to certain antibodies, software used in antibody development and apoptosis (programmed cell death) technology.

Other Agreements

Cytovance Biologics LLC

        In August 2007, we entered into an agreement with Cytovance Biologics LLC, or Cytovance, to develop a process for production of our huN901 antibody in accordance with current Good Manufacturing Practices, or cGMP, for potential use in IMGN901 clinical materials for pivotal trials and commercial applications. Under the terms of the agreement, we pay Cytovance incremental amounts for each step in the development process.

Laureate Pharma, Inc.

        In October 2007, we entered into an agreement with Laureate Pharma, Inc., or Laureate, to develop a process for cGMP production of our huC242 antibody for potential use in IMGN242 clinical materials for pivotal trials and commercial applications. Under the terms of the agreement, we pay Laureate incremental amounts for each step in the development process.

14

Table of Contents

Patents, Trademarks and Trade Secrets

        We seek patent protection for our proprietary technologies, product candidates, and related innovations in the U.S., Europe, Japan and elsewhere. Patents that have been issued to us in the U.S. include the following: claiming composition and use of certain maytansinoids; claiming conjugates composed of maytansinoids and cell-binding agents; claiming a process for the preparation of certain maytansinoids; claiming methods of preparation of conjugates composed of maytansinoids and cell-binding agents; and a method of antibody humanization. In many cases, we have received comparable patents outside the U.S.

        We have also submitted additional patent applications in the U.S., Europe, Japan, and elsewhere covering proprietary derivatives of cell-killing molecules, methods of attachment of such molecules to antibodies, TAP compounds, antibody compounds and use of some of these product candidates and inventions for certain diseases. We expect that our work will also lead to other patent applications. In all such cases, we will either be the assignee or owner of such patents or have an exclusive license to the technology covered by the patents. We cannot provide assurance, however, that the patent applications will issue as patents or that any patents, if issued, will provide us with adequate protection against competitors with respect to the covered products, technologies or processes.

        In addition, many of the processes and much of the know-how that are important to us depend upon the skills, knowledge and experience of our key scientific and technical personnel, which skills, knowledge and experience are not patentable. To protect our rights in these areas, we require that all employees, consultants, advisors and collaborators enter into confidentiality agreements with us. Further, we require that all employees enter into assignment of invention agreements as a condition of employment. We cannot provide assurance, however, that these agreements will provide adequate or any meaningful protection for our trade secrets, know-how or other proprietary information in the event of any unauthorized use or disclosure of such trade secrets, know-how or proprietary information. Further, in the absence of patent protection, we may be exposed to competitors who independently develop substantially equivalent technology or otherwise gain access to our trade secrets, know-how or other proprietary information.

Competition

        We focus on highly competitive areas of product development. Our competitors include major pharmaceutical companies and other biotechnology firms. Three of these companies, Wyeth, Seattle Genetics, Inc., and Medarex, Inc. have programs to attach a proprietary cell-killing small molecule to an antibody for targeted delivery to cancer cells. Pharmaceutical and biotechnology companies, as well as other institutions, also compete with us in recruiting highly qualified scientific personnel. Many competitors and potential competitors have substantially greater scientific, research and product development capabilities, as well as greater financial, marketing and human resources than we do. In addition, many specialized biotechnology firms have formed collaborations with large, established companies to support the research, development and commercialization of products that may be competitive with ours.

        In particular, competitive factors within the antibody and cancer therapeutic market include:

•

the safety and efficacy of products;



•

the timing of regulatory approval and commercial introduction;



•

special regulatory designation of products, such as Orphan Drug designation; and



•

the effectiveness of marketing, sales, and reimbursement efforts.

        Our competitive position depends on our ability to develop effective proprietary products, implement clinical development programs, production plans and marketing plans, including

15

Table of Contents

collaborations with other companies with greater marketing resources than ours, and to obtain patent protection and secure sufficient capital resources.

        Continuing development of conventional and targeted chemotherapeutics by large pharmaceutical companies and biotechnology companies may result in new compounds that may compete with our product candidates. In addition, antibodies developed by certain of these companies have been approved for use as cancer therapeutics. In the future, additional antibodies may compete with our product candidates. In addition, other companies have created or have programs to create potent cell-killing agents for attachment to antibodies. These companies may compete with us for technology out-license arrangements.

        Because of the acceptance of combination therapy for the treatment of cancer and the variety of genes and targets implicated in cancer incidence and progression, we believe that products resulting from applications of new technologies may be complementary to our own.

        Such new technologies include, but are not limited to:

•

the use of genomics technology to identify new gene-based targets for the development of anticancer drugs;



•

the use of high-throughput screening to identify and optimize lead compounds;



•

the use of gene therapy to deliver genes to regulate gene function; and



•

the use of therapeutic vaccines.

Regulatory Matters

        Our product candidates are regulated in the U.S. by the FDA in accordance with the U.S. Federal Food, Drug, and Cosmetic Act, as well as the Public Health Service Act. We expect that IMGN242, IMGN901, IMGN388 and other TAP compounds developed by us or our collaborators will be reviewed by the FDA's Center for Drug Evaluation and Research, or CDER. In addition, each drug manufacturer in the U.S. must be registered with the FDA.

        The steps required before a new drug may be marketed in the U.S. include:

(1)

Performance of preclinical laboratory, animal, and formulation studies;



(2)

The submission to the FDA of an IND application, which must become effective before clinical trials may commence;



(3)

The completion of adequate and well-controlled human clinical trials to establish the safety and efficacy of the drug;



(4)

The submission of a New Drug Application to and its acceptance by the FDA; and



(5)

FDA approval of the New Drug Application, including approval of product labeling and advertising.

        Even if we, or our collaborators, obtain regulatory approvals for our product candidates, our products and the facilities in which our products are manufactured are subject to continual review and periodic inspection. The FDA will require post-marketing reporting to monitor the safety of our products. Manufacturing establishments are subject to periodic inspections by the FDA and must comply with the FDA's cGMP regulations. In complying with cGMP regulations, manufacturers must expend funds, time and effort in the areas of production, quality control and recordkeeping to ensure full technical compliance. The FDA stringently applies regulatory standards for manufacturing.

16

Table of Contents

        The regulatory considerations that have potential impact on the future marketing of our product candidates are summarized below.

Clinical Trials Process

        Before a new drug may be sold in the U.S. and other countries, clinical trials of the product candidate must be conducted and the results submitted to the appropriate regulatory agencies for approval.

        In the U.S., these clinical trial programs generally involve a three-phase process. Typically, Phase I trials are conducted in healthy volunteers to determine the early side-effect profile and the pattern of drug distribution and metabolism. In Phase II, trials are conducted in groups of patients afflicted with the target disease to determine preliminary efficacy and optimal dosages and to expand the safety profile. In Phase III, large-scale comparative trials are conducted in patients with the target disease to provide sufficient data for the proof of efficacy and safety required by federal regulatory agencies. In the case of drugs for cancer and other life-threatening diseases, Phase I human testing usually is performed in patients with advanced disease rather than in healthy volunteers.

        We intend to conduct clinical trials not only in accordance with FDA regulations, but also within guidelines established by other applicable agencies and committees. Whether or not FDA approval has been obtained, approval of a product candidate by the comparable regulatory authorities of foreign countries must be obtained prior to the commencement of marketing of the product candidate in those countries. Regulatory approval in other countries is obtained through the various regulatory bodies governing pharmaceutical sales in those individual countries. We intend to rely on foreign licensees to obtain regulatory approvals to market our product candidates in foreign countries.

        Regulatory approval takes a number of years and involves the expenditure of substantial resources. Approval times also depend on a number of factors including, but not limited to, the severity of the disease in question, the availability of alternative treatments and the risks and benefits of the product candidate demonstrated in clinical trials.

Orphan Drug Designation

        Under the Orphan Drug Act, the FDA may grant orphan drug designation to a drug intended to treat a rare disease or condition, which is generally a disease or condition that affects fewer than 200,000 individuals in the U.S., or more than 200,000 individuals in the U.S. and for which there is no reasonable expectation that the cost of developing and making available in the U.S. a drug for this type of disease or condition will be recovered from sales in the U.S. for that drug. Orphan drug designation must be requested before submitting an NDA. After the FDA grants orphan drug designation, the identity of the therapeutic agent and its potential orphan use are disclosed publicly by the FDA. Orphan drug designation does not convey any advantage in or shorten the duration of the regulatory review and approval process.

        If a product that has orphan drug designation subsequently receives the first FDA approval for the disease for which it has such designation, the product is entitled to orphan product exclusivity, which means that the FDA may not approve any other applications to market the same drug for the same indication, except in very limited circumstances, for seven years. Orphan drug exclusivity, however, also could block the approval of one of our products for seven years if a competitor obtains approval of the same drug as defined by the FDA or if our product candidate is determined to be contained within the competitor's product for the same indication or disease.

        The FDA also administers a clinical research grants program, whereby researchers may compete for funding to conduct clinical trials to support the approval of drugs, biologics, medical devices, and medical foods for rare diseases and conditions. A product does not have to be designated as an orphan

17

Table of Contents

drug to be eligible for the grant program. An application for an orphan grant should propose one discrete clinical study to facilitate FDA approval of the product for a rare disease or condition. The study may address an unapproved new product or an unapproved new use for a product already on the market.

        We may pursue this designation with respect to product candidates intended for qualifying patient populations.

New Drugs for Serious or Life-Threatening Illnesses

        The FDA Modernization Act allows the designation of "Fast Track" status to expedite development of new drugs, including review and approvals, and is intended to speed the availability of new therapies to desperately ill patients. "Fast T