The following discussion of our financial condition and results of operations should be read in conjunction with the condensed consolidated financial statements and the related notes appearing elsewhere in this report.

Overview

We are a drug discovery and development company that is committed to leveraging our innovative signaling pathway drug technologies in seeking to develop next generation network-targeted cancer therapies. We are building upon our experience in modulating signaling pathways, including the Hedgehog signaling pathway, in our effort to develop network-targeted cancer therapies. We conduct our research and development programs both internally and through strategic collaborations.

Hedgehog Pathway Inhibitor Program (Erivedge™)

Erivedge™ (vismodegib) capsule. Our most advanced program is our Hedgehog pathway inhibitor program under collaboration with Genentech, Inc., a member of the Roche Group. The lead drug candidate being developed under this program is Erivedge, a first-in-class orally-administered small molecule Hedgehog pathway inhibitor, which is also referred to as vismodegib. Erivedge is designed to selectively inhibit signaling in the Hedgehog pathway by targeting a protein called Smoothened. The Hedgehog signaling pathway plays an important role in regulating proper growth and development in the early stages of life and becomes less active in adults. However, mutations in the pathway that reactivate Hedgehog signaling are seen in certain cancers, including basal cell carcinoma, or BCC. Abnormal signaling in the Hedgehog pathway is implicated in over 90% of BCC cases.

In January 2012, the U.S. Food and Drug Administration ("FDA") approved Genentech's new drug application, or NDA, for the Erivedge™ capsule for the treatment of adults with BCC that has spread to other parts of the body or that has come back after surgery or that their healthcare provider decides cannot be treated with surgery or radiation. Erivedge is being developed and commercialized by Roche and Genentech. As a result of the FDA's approval of Erivedge in this indication, we earned a $10,000,000 payment from Genentech and we are also entitled to receive royalties on future sales of the product. During the first quarter of 2012, we recognized the $10,000,000 milestone payment as license revenue for the three months ended March 31, 2012. In addition, we recorded research and development expenses related to the FDA's approval of Erivedge of $1,464,000 during the three months ended March 31, 2012 which represents our obligations to university licensors. Of this amount, $964,000 represents the fair value of a one-time issuance of an aggregate of 200,000 shares of our common stock to two university licensors in connection with the FDA-approval of Erivedge. The remaining $500,000 represents sublicense fees we paid to these same licensors, upon receipt of the $10,000,000 milestone payment in the first quarter of 2012.

We also recognized $271,000 of royalty revenue from Genentech's net sales of Erivedge during the quarter ended March 31, 2012. We recorded costs of royalty revenues of $114,000 during the three months ended March 31, 2012, including a $100,000 one-time cash payment paid to a university licensor upon the first commercial sale of Erivedge and $14,000 paid to two university licensors, which represents 5% of the royalties that we earned with respect to Erivedge during the first quarter of 2012.

During the fourth quarter of 2011, Roche submitted a Marketing Authorization Application, or MAA, for Erivedge to the European Medicines Agency, or EMA, for which we earned a $6,000,000 milestone payment. Roche has indicated that it anticipates potential EMA approval for Erivedge during the second half of 2012 or the first half of 2013. Roche has also filed new drug applications in 2012 for marketing registration with Australian, Canadian and Swiss health agencies seeking approval for Erivedge in advanced BCC. In May 2012 we earned a $4,000,000 milestone payment in connection with Roche's filing in Australia. Genentech's FDA approval and Roche's regulatory submissions in Europe, Australia, Canada and Switzerland are based on positive clinical data from ERIVANCE BCC/SHH4476g, a pivotal phase II study of Erivedge in patients with advanced BCC. We will receive additional milestone payments if Erivedge receives EMA marketing authorization or approval in Australia. We are also entitled to receive royalties on future net sales in all territories where Erivedge is sold.

In certain specified circumstances, the royalty rate applicable to Erivedge may be decreased to a low-to-mid single digit royalty, including when a competing product that binds to the same molecular target as Erivedge is approved by the applicable regulatory authority and is being sold in such country by a third party for use in the same indication as Erivedge.

Genentech is also conducting a separate phase II clinical trial of Erivedge in patients with operable nodular BCC, which is a less severe form of the disease and accounts for a significant percentage of the approximately two million BCCs diagnosed annually in the United States. This phase II trial is the first study to assess the ability of Erivedge to provide complete histological clearance of tumor, an important first step in determining the efficacy of Erivedge in less severe forms of BCC, where BCC lesions are generally treated surgically. This trial is designed to test Erivedge as a single-agent therapy in approximately 75 patients with operable nodular BCC in a US-based, open label, three-cohort clinical trial. Patients in the first and second cohorts receive a 150 mg daily oral dose of Erivedge for 12 weeks, while patients in the third cohort receive 16 weeks of daily dosing, with two eight-week dosing cycles surrounding a four-week period in which patients will not receive Erivedge. The primary outcome measure for the first and third cohorts is the rate of complete histological clearance of the target nodular BCC lesions at the time of tumor excision (which may occur up to 12 or 20 weeks, respectively, following initiation of treatment) while the primary outcome measure for the second cohort is the rate of durable complete clearance of target nodular BCC lesions at the time of excision (which may occur up to 36 weeks following initiation of treatment).

Data from the first cohort was published in April 2012 in the Journal of Investigative Dermatology and will be presented at the Society of Investigative Dermatology Annual Meeting in May 2012. This first cohort evaluated the safety and efficacy of 12 weeks of daily 150 mg dosing of Erivedge in 24 patients with newly diagnosed nodular, operable BCC. Patients then underwent Mohs surgery with independent pathology review. Clinical complete and partial responses were reported for 23 (96%) patients, including pathologically confirmed complete clearance in 10 (42%) patients. The most frequent adverse events (AEs) were similar to those observed in previous studies with Erivedge and included muscle spasms (79%), ageusia/dysgeusia (79%), alopecia (33%), fatigue (21%) and nausea (21%). Most AEs were Grade 1-2; seven patients (29%) reported Grade 3 AEs, including four patients with muscle spasm; no serious AEs were reported. Eight (33%) patients discontinued from the study, including two due to AEs. Accrual to cohorts two and three is ongoing with full study results expected in first half of 2013.

In addition to the BCC clinical trials being conducted directly by Genentech and Roche, Erivedge is also currently being tested in other cancers in trials under collaborative agreements between Genentech and either third-party investigators or the U.S. National Cancer Institute, or NCI.

Network-Targeted Cancer Programs

Our internal drug development efforts are focused on our network-targeted cancer programs, in which we are seeking to design single novel small molecule drug candidates that inhibit multiple signaling pathways that are believed to play roles in cancer cell proliferation. We refer to this approach as cancer network disruption. We believe that our approach of targeting multiple nodes in cancer signaling pathway networks may provide a better therapeutic effect than many of the cancer drugs currently marketed or in development which are designed to disrupt only one target.

CUDC-101. Our lead candidate from these programs is CUDC-101, a first-in-class small molecule compound designed to simultaneously target epidermal growth factor receptor, or EGFR, human epidermal growth factor receptor 2, or HER2, and histone deacetylase, or HDAC, all of which are validated cancer targets. A significant amount of our capital resources are focused on the ongoing clinical and preclinical development of this molecule. To date, we have completed a phase I dose escalation clinical trial of CUDC-101 in 25 patients with advanced, refractory solid tumors and a phase I expansion trial to test CUDC-101 in 46 patients with specific tumor types, including breast, gastric, head and neck, liver and non-small cell lung cancers. The phase I expansion trial was designed as an open-label study in which patients were treated with CUDC-101 at the maximum tolerated dose, which was determined in the phase I dose escalation study to be 275 milligrams per meter2. The primary objectives of this study were to compare the safety and tolerability of CUDC-101 in subjects with these specific advanced solid tumors when the drug was administered via one-hour intravenous infusion either on a five days per week schedule (one week on/one week off) or on a three days per week schedule (three weeks on/one week off).

In 2011, we initiated a phase I clinical trial of CUDC-101 in locally advanced head and neck cancer patients whose cancer is human papilloma virus, or HPV, negative. We have treated four patients in this trial as of May 3, 2012. The primary objectives of this study are to evaluate the safety and tolerability of CUDC-101 when administered in combination with the current standard-of-care of cisplatin, a chemotherapeutic drug, and radiation. Upon determination of the maximum tolerated dose and assuming the otherwise successful completion of this phase I trial, we intend to conduct a randomized phase II two-arm clinical trial in which head and neck cancer patients will receive cisplatin and radiation plus or minus CUDC-101. The phase II study would seek to evaluate whether the addition of CUDC-101 can improve the efficacy of cisplatin and radiation therapy in this patient population. We currently estimate initiating this phase II study in the first half of 2013.

We are also working on an oral formulation of CUDC-101, which we believe has the potential to make CUDC-101 more competitive in certain cancers such as non-small cell lung cancer or in other cancers where there are investigational or competing commercially available molecules that are orally administered. Pending the successful completion of ongoing formulation and preclinical development work, we intend to file an investigational new drug application, or IND, with the FDA and begin a phase I study of an oral formulation of CUDC-101 in the second half of 2012.

CUDC-907. In 2011, we selected development candidate CUDC-907, an orally bioavailable, network-targeted small molecule that is designed to inhibit phosphatidylinositol-3-kinase, or PI3K, and HDAC. Our scientists are developing CUDC-907 based on published and internally generated data demonstrating that HDAC and PI3K inhibitors have synergistic interaction in certain preclinical cancer models. We believe that this synergistic mechanism of cancer signaling network disruption, which demonstrated efficacy and a favorable safety profile in a number of preclinical xenograft models, could translate into clinical advantages over single targeted agents.

In November 2011, we entered into an agreement with The Leukemia & Lymphoma Society, or LLS, under which LLS will provide a portion of the funding of the development of CUDC-907 if we succeed in advancing this development candidate into a clinical trial for patients with B-cell lymphoma and multiple myeloma. Pending the successful completion of ongoing formulation and preclinical development work, we expect to file an IND with the FDA and begin a phase I study of CUDC-907 during the second half of 2012.

In addition to our development-stage programs, we continue to progress additional proprietary preclinical research programs and expect that we will select additional small molecule inhibitors from our preclinical portfolio in the future.

Hsp90 Program

Debio 0932. Our heat shock protein 90, or Hsp90, program is being developed by Debiopharm, a Swiss pharmaceutical development company, under an August 2009 license agreement between Curis and Debiopharm. The lead molecule under this license collaboration was designated Debio 0932 by Debiopharm. In April 2010, Debiopharm treated the first patient in a phase I clinical trial to evaluate the safety of Debio 0932 in patients with advanced solid tumors. In 2011, Debiopharm successfully advanced Debio 0932 through the dose escalation portion of the phase I study. Debio 0932 was generally well tolerated, with no evidence of ocular or liver toxicity, and showed promising signs of efficacy in patients with advanced solid tumors in this study. The recommended dose for further study was determined by Debiopharm to be 1000mg daily. Debiopharm has indicated that it expects to present data from this phase I study at the annual meeting of the American Society of Clinical Oncolgoy (ASCO) to be held in Chicago, Illinois in June 2012.

Debiopharm advanced Debio 0932 into the phase Ib expansion portion of the study in the beginning of 2012. The primary objectives of this portion of the study are to further assess the safety profile, pharmacokinetics and pharmacodynamics of Debio 0932 and to make a preliminary assessment of anti-tumor activity. Debiopharm expects that approximately 30 patients with advanced solid tumors will be treated in this portion of the study, including patients with non-small cell lung cancer (NSCLC).

Debiopharm has also indicated that it expects to initiate a combination Phase I/II study in NSCLC patients in the second quarter of 2012. We are eligible for our next milestone payment under our license agreement if and when Debiopharm treats its fifth patient in a phase II clinical trial, assuming that Debiopharm advances Debio 0932 into phase II clinical testing. We currently anticipate that phase II testing could commence in the first half of 2013.

Liquidity

Since our inception, we have funded our operations primarily through license fees, contingent cash payments, research and development funding from our corporate collaborators, the private and public placement of our equity securities, debt financings and the monetization of certain royalty rights. We have never been profitable on an annual basis and have an accumulated deficit of $729,862,000 as of March 31, 2012. We expect to continue to incur significant costs and expenses as we seek to advance our research and development programs. We expect that we will incur significant operating losses for the next several years as we devote substantially all of our resources to our research and development programs. Although Genentech recently received FDA approval to market Erivedge, the amount of future sales and resulting royalty revenue payable to us are highly uncertain. We will need to generate significant revenues to achieve profitability and do not expect to achieve profitability in the foreseeable future, if at all. We anticipate that existing capital resources as of March 31, 2012 should enable us to maintain current and planned operations into the first half of 2014. We believe that near term key drivers to our success will include:

• Genentech's ability to successfully scale up the commercialization of Erivedge in the US;

• Debiopharm's ability to advance Debio 0932 into later stages of clinical development;

• our ability to successfully plan, finance and complete current and planned clinical trials for CUDC-101 and advance CUDC-101 into later stages of clinical development in indications other than head and neck cancers;

• our ability to successfully advance early-stage development candidates, such as CUDC-907 into clinical testing;

• our ability to successfully enter into one or more material licenses or collaboration agreements for our proprietary drug candidates; and

• our ability to advance the research of other small molecule cancer drug candidates that we are developing under our proprietary pipeline of network-targeted cancer programs.

In the longer term, a key driver to our success will be our ability, and the ability of any current or future collaborator or licensee, to successfully commercialize drugs other than Erivedge based upon our proprietary technologies.

Collaboration Agreements

We are currently a party to a June 2003 collaboration with Genentech relating to our Hedgehog pathway inhibitor technologies, and an August 2009 license agreement with Debiopharm relating to our Hsp90 inhibitor technology. Our past and current collaborations have generally provided for research, development and commercialization programs to be wholly or majority-funded by our collaborators and provide us with the opportunity to receive additional contingent cash payments if specified development and regulatory approval objectives are achieved, as well as royalty payments upon the successful commercialization of any products based upon the collaborations. We are currently not receiving any research funding and we do not expect to receive such funding in the future from Genentech or Debiopharm under our current agreements with these parties. Under our collaborations with Genentech, we currently expect to incur only costs related to the maintenance of licenses, including sublicense payments due upon milestone payments and any royalties we receive, as well as patent-related expenses. As a result of our licensing agreements with various universities, we are also obligated to make payments to these university licensors when we receive certain payments from Genentech. As of March 31, 2012, we have incurred an aggregate of approximately $2,250,000 related to ongoing agreements, of which $2,200,000 relates to payments that we have or will receive from Genentech. As we receive additional milestone payments from Genentech upon EMA/AUS approval, if achieved, we will be obligated to pay additional sublicense fees to these licensors, as well as fees related to any royalties received from the sale of Erivedge. In addition, we were obligated to issue 200,000 shares of our common stock to certain licensees upon FDA approval of Erivedge that represented $964,000 in expense during the first quarter of 2012. We do not expect to incur any material costs related to our Hsp90 technologies under development by Debiopharm under our August 2009 license agreement with Debiopharm.

Financial Operations Overview

General. Our future operating results will largely depend on the magnitude of payments from our current and potential future corporate collaborators and the progress of drug candidates currently in our research and development pipeline. The results of our operations will vary significantly from year to year and quarter to quarter and depend on, among other factors, the timing of our entry into new collaborations, if any, the timing of the receipt of payments, if any, from new or existing collaborators and the cost and outcome of any preclinical development or clinical trials then being conducted. We anticipate that existing capital resources as of March 31, 2012 should enable us to maintain current and planned operations into the first half of 2014. Our ability to continue funding our planned operations into and beyond the first half of 2014 is dependent on future contingent payments that we may receive from Debiopharm or Genentech upon the achievement of development and regulatory approval objectives, our ability to manage our expenses and our ability to raise additional funds through additional corporate collaborations, equity or debt financings, or from other sources of financing.

A discussion of certain risks and uncertainties that could affect our liquidity, capital requirements and ability to raise additional funds is set forth under "Part II, Item 1A-Risk Factors."

Revenue. We do not expect to generate any revenues from our direct sale of products for several years, if ever. Substantially all of our revenues to date have been derived from license fees, research and development payments, and other amounts that we have received from our strategic collaborators and licensees. We expect to recognize royalty revenue in future quarters from Genentech's sales of Erivedge in the U.S. and in other markets where Genentech and Roche successfully obtain marketing approval.

We could receive additional milestone payments from Genentech and Debiopharm, provided the respective programs meet contractually-specified development and regulatory objectives. For example, we earned a $10,000,000 milestone payment from Genentech in January 2012 upon FDA approval of Erivedge. In addition, Erivedge is currently being reviewed for potential marketing approval by the EMA in European territories as well as by Australian, Canadian and Swiss health authorities. We are eligible to receive additional milestone revenue should Erivedge receive approval by the EMA and/or Australian health authorities, and we are also eligible to receive royalties on net sales of Erivedge in all territories where Erivedge is sold.

We currently receive no research funding for our programs under our collaborations with Genentech and Debiopharm and we do not expect to receive such funding in the future under these collaborations. Accordingly, our only source of revenues and/or cash flows from operations for the foreseeable future will be up-front license payments and funded research and development that we may receive under new collaboration agreements, if any, contingent cash payments for the achievement of clinical development and regulatory objectives, if any are met, under new collaborations or our existing collaborations with Genentech and Debiopharm and royalty payments that are contingent upon the successful commercialization of any products based upon these collaborations. Our ability to enter into new collaborations and our receipt of additional payments under our existing collaborations with Genentech and Debiopharm cannot be assured, nor can we predict the timing of any such arrangements or payments, as the case may be.

Research and Development. Research and development expense consists of costs incurred to discover, research and develop our drug candidates. These expenses consist primarily of: (1) salaries and related expenses for personnel including stock-based compensation expense; (2) outside service costs including clinical research organizations, medicinal chemistry; (3) sublicense payments; and
(4) the costs of supplies and reagents, consulting, and occupancy and depreciation charges. We expense research and development costs as incurred. We are currently incurring research and development expenses under our Hedgehog pathway inhibitor collaboration with Genentech related to the maintenance of third-party licenses to certain background technologies. For each contingent payment, if any, and royalty payments that we receive from Genentech under our Hedgehog pathway inhibitor collaboration, we are obligated to make payments totaling 5% of these amounts to certain third-party university licensors. We will recognize these payments as research and development expense in our financial statements as the related expense is incurred.

Our research and development programs, both internal and under collaboration, are summarized in the following table:

Product Candidate Primary Disease Collaborator/Licensee Status Hedgehog Pathway Inhibitor
- Erivedge (vismodegib) Advanced BCC Genentech FDA approved; Regulatory submissions pending in EU, AUS, CAD, SWISS review

- Erivedge (vismodegib) Operable Nodular BCC Genentech Phase II

Network-targeted Cancer Programs
-   CUDC-101 intravenous       Cancer                   Internal development    Phase I expansion
    formulation (EGFR, Her2,
    HDAC inhibitor)
-   CUDC-101 intravenous       Locally advanced HPV-    Internal development    Phase I
    formulation (EGFR, Her2,   head and neck cancer
    HDAC inhibitor)
-   CUDC-101 oral              Cancer                   Internal development    Development candidate
    formulation (EGFR, Her2,
    HDAC inhibitor)
-   CUDC-907 (PI3K, HDAC       Cancer                   Internal development    Development candidate
    inhibitor)
-   Other network-targeted     Cancer                   Internal development    Preclinical
    cancer programs
-   Debio 0932 (formerly       Cancer                   Debiopharm              Phase Ib
    CUDC-305) (Hsp90
    inhibitor)

In the chart above, "Phase II" means that Genentech is currently treating human patients in a phase II clinical trial, the primary objective of which is a therapeutic response in the patient population. "Phase I expansion" means that we have completed treating human patients with specific tumor types in an extension of our phase I dose escalation trial, at the maximum tolerated dose from such trial, the principal purpose of which is to evaluate the safety and tolerability of the compound being tested, and that we are currently compiling final study data for future presentation at a medical conference.

"Phase Ib" means that Debiopharm is further assessing the safety profile, pharmacokinetics and pharmacodynamics of Debio 0932 at the recommended phase II dose level, and seeking to make a preliminary assessment of anti-tumor activity in patients with advanced solid tumors. "Phase I" means that we are currently treating human patients in a phase I clinical trial with CUDC-101, the principal purpose of which is to evaluate the safety and tolerability of the compound being tested. "Development candidate" means that we have selected a single lead candidate for potential future clinical development and are seeking to complete the relevant safety, toxicology, and other studies required to submit an IND application with the FDA seeking to commence a phase I clinical trial based on our testing in several preclinical models of human disease of various compounds from a particular compound class. "Preclinical" means that we are seeking to obtain evidence of therapeutic efficacy and safety in preclinical models of human disease of one or more compounds within a particular class of drug candidates.

Because of the early stages of development of these programs, our ability and that of our collaborators and licensees to successfully complete preclinical studies and clinical trials of these drug candidates, and the timing of completion of such programs, is highly uncertain. There are numerous risks and uncertainties associated with developing drugs which may affect our and our collaborators' future results, including:

• the scope, quality of data, rate of progress and cost of clinical trials and other research and development activities undertaken by us or our collaborators;

. . .