Treating solid tumors with cancer immunotherapy (CIT) can result in unconventional responses and overall survival (OS) benefits that are not adequately captured by Response Evaluation Criteria In Solid Tumors (RECIST) v1.1. We describe immune-modified RECIST (imRECIST) criteria, designed to better capture CIT responses.

Atezolizumab data from clinical trials in non–small-cell lung cancer, metastatic urothelial carcinoma, renal cell carcinoma, and melanoma were evaluated. Modifications to imRECIST versus RECIST v1.1 included allowance for best overall response after progressive disease (PD) and changes in PD definitions per new lesions (NLs) and nontarget lesions. imRECIST progression-free survival (PFS) did not count initial PD as an event if the subsequent scan showed disease control. OS was evaluated using conditional landmarks in patients whose PFS differed by imRECIST versus RECIST v1.1.

The best overall response was 1% to 2% greater, the disease control rate was 8% to 13% greater, and the median PFS was 0.5 to 1.5 months longer per imRECIST versus RECIST v1.1. Extension of imRECIST PFS versus RECIST v1.1 PFS was associated with longer or similar OS. Patterns of progression analysis revealed that patients who developed NLs without target lesion (TL) progression had a similar or shorter OS compared with patients with RECIST v1.1 TL progression. Patients infrequently experienced a spike pattern (TLs increase, then decrease) but had longer OS than patients without TL reversion.

Evaluation of PFS and patterns of response and progression revealed that allowance for TL reversion from PD per imRECIST may better identify patients with OS benefit. Progression defined by the isolated appearance of NLs, however, is not associated with longer OS. These results may inform additional modifications to radiographic criteria (including imRECIST) to better reflect efficacy with CIT agents.

Response Evaluation Criteria In Solid Tumors (RECIST) v1.1 end points, originally developed to evaluate the benefit of chemotherapeutic and targeted agents, are accepted measures of clinical efficacy in advanced solid malignancies.1 Recently, cancer immunotherapy (CIT) has shifted the treatment paradigm for several cancers.2,3 In addition to classic response patterns defined by RECIST, other patterns can manifest with CIT, with frequency varying by tumor type. These patterns include an initial transient increase in tumor burden before response4,5 and/or appearance of new lesions (NLs) in patients with responding baseline lesions.6-8 This may be due to anticancer immune activity leading to tumor inflammation or the dynamic nature of immunity, which may depend on tumor, host, or environmental factors.9 Experience with immune checkpoint inhibitors has shown that the overall survival (OS) benefit with CIT is often not fully reflected in RECIST v1.1–based progression-free survival (PFS) or overall response rate (ORR).10-15 Therefore, RECIST v1.1–based end points can underestimate the clinical benefit of CIT, reducing their reliability as surrogate efficacy measures. Accordingly, extensive efforts have been and continue to be made to develop alternate radiographic criteria to measure efficacy of CIT.

Immune-related response criteria (irRC) were developed based on experience with ipilimumab (anti–cytotoxic T-cell lymphocyte-4) in melanoma to better capture the response to CIT per changes in tumor biology and the long-term effects of CIT on solid tumors and to enable additional adaptations as the field evolved.6,7,16 irRC use WHO criteria with bidimensional tumor measurement, with alternate definitions of progressive disease (PD) on the basis of inclusion of NLs in the overall calculated tumor burden, and allow for response after initial progression. Additional adaptations to the RECIST v1.1 framework with unidimensional measurements have also been developed and applied. Adaptation of irRC to a unidimensional framework has yielded best response results comparable to bidimensional irRC with higher reproducibility.8,17 Recently, the RECIST working group published modified RECIST 1.1 for immune-based therapeutics (iRECIST) criteria to encourage consistency in immune-based radiographic data capture and analysis.18

irRC and RECIST-based adaptations have been widely applied to ORR in studies of programmed death-ligand 1 (PD-L1) and programmed death-1 (PD-1) inhibitors. However, despite the potential for a profound impact on end point selection, the application of immune-based criteria to PFS analyses has only been partially explored. Moreover, evaluation of these criteria to predict OS has been limited. Survival analyses according to differences in best overall response (BOR) demonstrated that ipilimumab-treated patients with melanoma who had PD by WHO criteria but stable disease (SD)/partial response (PR) by irRC had survival comparable to that in patients with SD/PR/complete response (CR) per WHO criteria.6 Similar results were seen for pembrolizumab in melanoma.7 Although these analyses highlight OS association methods that can be used to evaluate immune-based criteria performance as end points, additional systematic examination of CIT-specific unconventional patterns and associated OS in other tumor types remains to be performed.

We describe immune-modified RECIST (imRECIST) criteria, which include adaptations for CIT response patterns and definitions applicable to PFS analyses, and summarize results using imRECIST for evaluation of atezolizumab (anti–PD-L1).11,19 In addition, we examine whether differences in PFS per imRECIST versus RECIST v1.1 translate into OS benefit from atezolizumab. Lastly, we investigate the association of tumor response patterns with OS by imRECIST versus RECIST v1.1. These analyses reveal aspects of imRECIST that seem to predict OS better than RECIST v1.1 and aspects needing additional refinement to improve the ability to predict clinical benefit.

Study Design and Patients

We used patient data from phase I and II clinical trials evaluating atezolizumab monotherapy across indications. To evaluate the ORR and PFS per imRECIST, we examined non–small-cell lung cancer (NSCLC) data from cohorts 2 and 3 of the BIRCH (NCT02031458)20 and POPLAR (NCT01903993) phase 2 clinical trials14 and metastatic urothelial carcinoma (mUC) data from cohort 2 of the IMvigor210 (NCT02108652)21 phase 2 clinical trial. To evaluate progression patterns, we examined additional data from the renal cell carcinoma (RCC) and melanoma cohorts of the PCD4989g trial (NCT01375842).11,22

Treatment and Radiologic Assessments

Patients received intravenous atezolizumab at doses of up to 20 mg/kg (PCD4989g) or at 1,200 mg (BIRCH, POPLAR, IMvigor210) every 3 weeks. Patients in the POPLAR control arm received intravenous docetaxel at 75 mg/m2 every 3 weeks. Atezolizumab was administered until loss of investigator-assessed clinical benefit in the relevant trial cohorts. Patients who progressed per RECIST v1.1 could continue atezolizumab if they were experiencing investigator-assessed clinical benefit and did not have performance status decline, signs, or symptoms of unequivocal PD or PD at sensitive sites.

In BIRCH, POPLAR, and IMvigor210, investigators assessed tumors per RECIST v1.1 and imRECIST every 6 weeks, with reduced frequency after 9 to 12 months. In PCD4989g, investigators assessed tumors every 6 weeks for 24 weeks and every 12 weeks thereafter per RECIST v1.1 and irRC because the study was initiated before imRECIST criteria were developed. Therefore, data from PCD4989g were analyzed only for patterns of response and progression.

The imRECIST Criteria

The imRECIST criteria were developed initially for implementation in atezolizumab studies (Data Supplement).23 These criteria are modifications of the RECIST v1.1 system on the basis of the principles of irRC (Table 1).1,6,7 Briefly, imRECIST allows for collection of additional scans and for BOR to occur after radiologic PD assessment(s) in patients continuing treatment (Table 1; Data Supplement). NLs are added to the total tumor burden along with the sum of the target lesions (TLs) when measurable; when not measurable, they are not factored into the PD assessment. In addition, progression in nontarget lesions (NTLs) does not define PD. For analysis of imRECIST-defined PFS (imPFS), imRECIST PD or death is considered an event; however, an imRECIST PD is not considered an imPFS event if the time point response at the subsequent scan (≥ 4 weeks later) is imRECIST SD/PR/CR. An imRECIST PD followed by no additional assessments is considered an imPFS event (Data Supplement).


Table 1. Comparison of imRECIST With RECIST v1.1 and irRC

OS per occurrence of RECIST v1.1 PFS versus imPFS events was evaluated using conditional landmarks to account for survivor bias (guaranteed time bias).24 Among all patients who were alive ≥ 90 days and ≥ 180 days after enrollment, we analyzed OS of those with a RECIST v1.1 PFS event only and those with both RECIST v1.1 and imPFS events within the respective landmark times. We selected these landmarks because of the proximity to the median PFS by RECIST v1.1 observed with atezolizumab across the included indications. Eight-month landmark survival was assessed to further characterize OS curves.

Response and Progression Patterns

Three progression patterns were defined per lesion type (TL/NL/NTL; Data Supplement): (1) first progression in TLs (may include simultaneous NL or NTL progression); (2) first progression in NLs but not TLs (may include NTL progression); and (3) first progression in NTLs only. To further evaluate patients with the first pattern, we also examined those who had a TL reversion response pattern, defined as an increase of ≥ 20% in TL tumor burden and subsequent return to < +20%, relative to the original baseline or nadir (Data Supplement).

OS from enrollment was evaluated among patient subgroups on the basis of previously described pattern definitions. For the TL reversion pattern and the complementary pattern of TL progression without reversion to < +20%, a conditional landmark analysis was used to account for survivor bias. Patients who experienced PD within 90 days of enrollment and were alive at day 90 were included, and OS was evaluated according to whether reversion of TLs occurred at any time. OS for the previously defined progression patterns was determined without conditional landmarks; medians and 8-month survival were summarized.

Statistical Analysis

Kaplan-Meier methodology was used to construct survival curves, estimate median OS and PFS, and provide landmark analyses. Brookmeyer-Crowley methodology was used to construct 95% CIs for medians. Normal approximation was used for the 8-month landmark CI. The Clopper-Pearson method was used to produce 95% CIs for response rates.

BOR was examined for patients with NSCLC and mUC in BIRCH, POPLAR, and IMvigor210, using RECIST v1.1 and imRECIST. The ORR was 15% to 20% across studies, with a 1% to 2% increase per imRECIST versus RECIST v1.1 (Table 2). The disease control rate (DCR; defined as CR plus PR plus SD) was 8% to 13% higher with imRECIST than with RECIST v1.1 (Table 2). Observed increases in disease control rate by imRECIST were largely driven by a higher SD rate, potentially reflecting the underlying patterns of response to anti–PD-L1 therapy (Data Supplement). Alternatively, or in addition, the difference in PD definitions (Table 1) may result in higher SD rates.


Table 2. Response and PFS by imRECIST Versus RECIST v1.1

Next, we compared PFS per RECIST v1.1 and imRECIST (Table 2) and observed that imPFS was 0.5 to 1.5 months longer than RECIST v1.1 PFS. This difference was more pronounced in patients with NSCLC than in those with mUC. The longer imPFS is consistent with the increased SD rate per imRECIST and may be explained by the reasons for higher SD rates described previously and/or negation of an imRECIST PD event by subsequent SD/PR/CR assessment.

To evaluate whether differences in imPFS versus RECIST v1.1 PFS were associated with OS in atezolizumab-treated patients, we examined OS according to PFS event timing (ie, whether PFS was the same or longer per imRECIST versus RECIST v1.1). Among all patients, 36% to 40% were alive and had RECIST v1.1 PD within 90 days of enrollment (Table 3). Of these patients, 66% to 76% had PD by both RECIST v1.1 and imRECIST, and the remainder had PD by RECIST v1.1 only. In BIRCH and POPLAR (atezolizumab-arm), median OS was 4.0 and 1.4 months longer, respectively, in patients with PD by RECIST v1.1 only than in those with PD by RECIST v1.1 and imRECIST (Table 3; Figs 1A and 1B), and 8-month OS rates were 9% and 28% higher, respectively (Table 3). No differences were seen in median OS or 8-month survival for IMvigor210 patients (Table 3; Fig 1C). Results using a 180-day landmark were similar for POPLAR, and medians were not estimable for BIRCH. However, a 4.4-month longer median OS was observed for IMvigor210 patients with PD by RECIST v1.1 only versus those with PD by RECIST v1.1 and imRECIST (Data Supplement). Notably, a higher proportion (72% to 100%) of patients who either had PD by RECIST v1.1 and either later (> 7 days) or no PD by imRECIST continued treatment beyond progression, as allowed per protocol versus those who had simultaneous (± 7 days) PD per RECIST v1.1 and imRECIST (40% to 50%).


Table 3. Frequency of PD Events and OS According to the Timing of Occurrence of PFS Events per RECIST v1.1 Compared With imRECIST Using a 90-Day Conditional Landmark in Patients Treated With Atezolizumab

To investigate the impact of key differences between RECIST v1.1 and imRECIST, we analyzed changes in tumor patterns observed with atezolizumab and their association with OS in BIRCH, POPLAR, IMvigor210, and PCD4989g (RCC and melanoma cohorts). First, we examined the frequency of and OS associated with NL and NTL progression relative to TL progression. The incidence of PD in TLs, which could include NL and/or NTL PD, was highest (48% to 61%; Table 4). Development of NLs without TL PD was common (32% to 41%; Table 4; Data Supplement), and progression in NTLs alone was less frequent (6% to 13%; Table 4). Notably, the median and 8-month OS in patients who developed NLs without TL PD was similar to or lower than that in patients who had PD in TLs (Table 4; Data Supplement). OS in patients with progression in NTLs alone was variable versus in patients with TL PD (Table 4). For comparison, we evaluated patterns and OS in docetaxel-treated patients in the POPLAR control arm (Table 4). The pattern frequencies were similar to those in the atezolizumab arm, and the NL and NTL patterns were both associated with lower OS than the TL pattern, similar to trends seen in atezolizumab-treated patients.


Table 4. Frequency and OS Associated With PD Due to Initial Progression in Target Lesions, New Lesions, and Nontarget Lesions in Patients Treated With Atezolizumab

Next, we examined the frequency of TL reversion, a response pattern defined differently per imRECIST versus RECIST v1.1 and the associated OS. Similar proportions of patients who had either initial progression relative to the baseline or initial decrease followed by 20% increase from nadir were observed among those who subsequently had reductions in tumor burden (Table 5; Data Supplement). Overall, TL reversion from PD (at any time) to non-PD occurred in 4% to 8% of patients across indications (Table 5), consistent with evaluation per irRC in melanoma.7 The rate of TL reversion in patients who had initial PD within 90 days of atezolizumab treatment onset was much smaller (1.5% to 3%), with the exception of melanoma (7%). Across indications, median OS was longer and 8-month survival was higher in patients with TL reversion post-PD versus without reversion, except those with melanoma, in whom 8-month OS was shorter (Table 5; Data Supplement).


Table 5. Frequency and OS Associated With Target Lesion Reversion Pattern in Patients Treated With Atezolizumab

With widespread development of new CIT agents for the treatment of a broad array of cancer types, there is a need to adapt tumor assessment criteria to accommodate unconventional responses and thereby allow robust evaluation of efficacy benefits with surrogate end points (ORR, PFS) that typically mature well in advance of OS. Numerous trials to date have shown that these efficacy measures often underestimate survival benefit when RECIST v1.1 criteria are applied.13-15 Work to address this discordance began with development of the irRC6,7 and has been extended here with imRECIST, a unidimensional measurement system on the basis of the RECIST v1.1 framework. This work examined data from atezolizumab-treated patients with NSCLC, mUC, RCC, and melanoma (1,082 total), providing a rich data set for evaluating and refining imRECIST criteria. imRECIST includes key principles of irRC applied with modification to RECIST v1.1; use of unidimensional criteria in imRECIST may reduce measurement variability and thereby improve reproducibility of response assessments.8,25-30

Evaluation of second-line (2L)+ NSCLC and mUC atezolizumab study data revealed similar ORRs by imRECIST and RECIST v1.1, consistent with observations with other anti–PD-L1/PD-1 agents across indications using immune-related criteria.7,31-34 This suggests that the rates of pseudoprogression, when defined as progression followed by response, are generally low for this class of CIT.7,34 The low pseudoprogression frequency suggests that disease growth stabilization/slowing may be as relevant as responses when measuring efficacy by radiographic changes; this would help account for observed OS efficacy not always mirroring high response rates. This was particularly evident for NSCLC in PD-L1–unselected populations treated with atezolizumab.14,15 For these reasons, imPFS methodology was developed and applied to atezolizumab studies. Medians using imPFS were longer than PFS by RECIST v1.1. Patients without imRECIST but with RECIST v1.1 progression as of a 90-day OS landmark showed the same or longer median survival than those who had progression by both criteria, suggesting imPFS may better reflect survival benefit for some patients. It is noteworthy that patients with progression per RECIST v1.1 only may have favorable clinical status, making them eligible per protocol for continued treatment with atezolizumab postprogression. However, this only reinforces the need for alternate radiographic criteria that can better reflect the improved clinical outcome for these patients. Taken together, this methodology for analyzing OS by differential PFS may be an important new tool for evaluating modifications to tumor response criteria.

Exploration of the response and progression patterns that represent key differences between imRECIST and RECIST v1.1 for several indications, along with the OS associated with these patterns, revealed several important findings. First, as with the classic pseudoprogression pattern, the frequency of patients experiencing TL PD followed by reversion to SD/response is relatively low; however, these patients had longer OS than patients without TL PD reversion, suggesting that the imRECIST provision for negation of a PD event if lesion burden subsequently improves may result in better prediction of OS benefit. Second, patients having initial progression in NLs but not TLs were common (20% to 34% of patients) but tended to have similar or shorter OS than those with initial PD in TLs. In evaluating a non-CIT comparator for context, a similar OS trend per these patterns was seen for docetaxel-treated patients with NSCLC. These data suggest that NLs remain an important component of PD definitions of immune-modified radiographic criteria (and of RECIST v1.1 for nonimmunotherapies) and that the imRECIST treatment of NLs may need additional examination and adjustment in future updates. Third, patients who had initial PD in NTLs only were limited in number, with mixed OS trends relative to those with TL PD. It is not clear whether the imRECIST and irRC considerations of measurable lesions only in the assessment of PD result in better or worse prediction of OS relative to the provision for unequivocal progression as part of RECIST v1.1. Of note, the recently developed iRECIST guideline includes a provision for unequivocal progression of NTLs in the assessment of a time point response of unconfirmed PD.18 Together, these results suggest that some aspects of imRECIST are important for better capturing clinical benefit with CIT, whereas others need further modification.

Although these results provide an important evaluation of imRECIST data in atezolizumab trials and the association of trends observed with OS, there are caveats in generalizing these results, and challenges remain for advancing the broad utility of these criteria. First, it is often nonstandard to continue treatment and RECIST evaluation in patients experiencing PD on nonimmunotherapies, making application of imRECIST criteria to both arms of many randomized trials difficult. There may also be differences in the patterns of responses in the same patient population between different CIT types (anti–cytotoxic T-cell lymphocyte-4 versus anti–PD-L1/PD-1, anti–PD-L1 versus anti–PD-1) and combinations. In addition, different response patterns might occur with the same agent across tumor types. To date, no relationship has been detected between PD-L1 expression levels on tumor cells or tumor-infiltrating immune cells and pseudoprogression as defined by irRC.7

The imPFS analyses presented here, along with methodologies for examining OS by different progression patterns, may serve as a framework to further refine imRECIST. For example, the prevalence of nonclassic response patterns may vary by lesion locations, timing relative to initiation of treatment, or NLs that subsequently regress. Lymph nodes, given their importance to the immune response, may show distinct patterns and require different rules for evaluation relative to other lesion types. In addition, observations of rapid growth, which have been reported in some patients receiving PD-L1/PD-1 inhibitor therapy, can also be considered.35,36

As investigative studies of different CITs across more patients with additional tumor types are conducted, we expect additional response patterns to emerge. Comparison of immunotherapies in both arms of randomized studies will provide opportunities for validating imRECIST criteria in prospective clinical trials. Of note, an international effort by the iRECIST group comparing data sets from larger sets of trials and multiple CIT agents is under way. Lessons learned from imRECIST evaluations may also provide valuable insights that may be applied to future iterations of the recently published iRECIST criteria.18

© 2018 by American Society of Clinical Oncology

Supported by F. Hoffmann-La Roche.

Presented in part at the American Society of Clinical Oncology Annual Meeting, Chicago, IL, June 3-7, 2016.

See accompanying Editorial on page 835

Conception and design: F. Stephen Hodi, Marcus Ballinger, Benjamin Lyons, Mizuki Nishino, Josep Tabernero, Thomas Powles, Axel Hoos, Chris McKenna, Ina Rhee, Gregg Fine, Nathan Winslow, Daniel S. Chen, Jedd D. Wolchok

Provision of study materials or patients: Josep Tabernero

Collection and assembly of data: F. Stephen Hodi, Marcus Ballinger, Benjamin Lyons, Thomas Powles, David Smith, Gregg Fine, Daniel S. Chen

Data analysis and interpretation: F. Stephen Hodi, Marcus Ballinger, Benjamin Lyons, Jean-Charles Soria, Mizuki Nishino, Josep Tabernero, Thomas Powles, Chris McKenna, Ulrich Beyer, Ina Rhee, Gregg Fine, Daniel S. Chen

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

Immune-Modified Response Evaluation Criteria In Solid Tumors (imRECIST): Refining Guidelines to Assess the Clinical Benefit of Cancer Immunotherapy

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to or

F. Stephen Hodi

Employment: Dana-Farber Cancer Institute

Consulting or Advisory Role: Merck Sharp & Dohme, Novartis, Genentech/Roche, Amgen, EMD Serono, Celldex, Bristol-Myers Squibb

Research Funding: Bristol-Myers Squibb (Inst), Merck Sharp & Dohme (Inst), Genentech/Roche (Inst), Novartis (Inst)

Patents, Royalties, Other Intellectual Property: Patent pending royalties received on MICA-related disorders application to institution per institutional IP policy

Travel, Accommodations, Expenses: Novartis, Bristol-Myers Squibb

Other Relationship: Bristol-Myers Squibb, Genentech/Roche

Marcus Ballinger

Employment: Genentech/Roche

Stock or Other Ownership: Roche/Genentech

Benjamin Lyons

Employment: Genentech/Roche

Stock or Other Ownership: Genentech/Roche, Johnson & Johnson

Jean-Charles Soria

Honoraria: Roche

Mizuki Nishino

Honoraria: Bayer Yakuhin

Consulting or Advisory Role: Toshiba, WorldCare Clinical

Research Funding: Merck (Inst), Toshiba (Inst), AstraZeneca (Inst)

Josep Tabernero

Consulting or Advisory Role: Amgen, Bayer, Boehringer Ingelheim, Celgene, Chugai Pharma, Eli Lilly, MSD, Merck Serono, Novartis, Pfizer, Roche, Sanofi, Symphogen, Taiho Pharmaceutical, Takeda

Thomas Powles

Honoraria: Exelixis

Consulting or Advisory Role: Genentech/Roche, Bristol-Myers Squibb, Merck, Novartis, AstraZeneca

Research Funding: AstraZeneca/MedImmune, Roche/Genentech

David Smith

Employment: US Oncology

Leadership: US Oncology

Consulting or Advisory Role: Takeda

Research Funding: US Oncology (Inst)

Travel, Accommodations, Expenses: US Oncology

Axel Hoos

Employment: GlaxoSmithKline

Leadership: Imugene

Stock or Other Ownership: GlaxoSmithKline, Imugene

Chris McKenna

Employment: Genentech/Roche

Stock or Other Ownership: Roche/Genentech

Travel, Accommodations, Expenses: Roche/Genentech

Ulrich Beyer

Employment: Roche

Ina Rhee

Employment: Genentech/Roche

Stock or Other Ownership: Roche/Genentech

Patents, Royalties, Other Intellectual Property: Roche

Gregg Fine

Employment: Genentech

Stock or Other Ownership: Genentech, Foundation Medicine

Nathan Winslow

Employment: Genentech/Roche

Stock or Other Ownership: Roche/Genentech

Daniel S. Chen

Employment: Genentech/Roche

Leadership: Genentech/Roche

Stock or Other Ownership: Genentech/Roche

Travel, Accommodations, Expenses: Genentech/Roche

Jedd D. Wolchok

Stock or Other Ownership: Potenza Therapeutics, Tizona Therapeutics, Serametrix, Adaptive Biotechnologies, Trieza Therapeutics, BeiGene, Imvaq Therapeutics

Consulting or Advisory Role: Bristol-Myers Squibb, Merck, MedImmune, Polynoma, Polaris, Genentech, F-Star, BeiGene, Sellas Life Sciences, Eli Lilly, Potenza Therapeutics, Tizona Therapeutics, Amgen, Chugai Pharma, Ascentage Pharma, Northern Biologics, Janssen Oncology

Research Funding: Bristol-Myers Squibb (Inst), Merck (Inst), Genentech/Roche (Inst), MedImmune (Inst)

Patents, Royalties, Other Intellectual Property: I am a coinventor on an issued patent for DNA vaccines for treatment of cancer in companion animals; I am a coinventor on a patent for use of oncolytic Newcastle disease virus

Travel, Accommodations, Expenses: Bristol-Myers Squibb, Chugai Pharma, Roche, Janssen


We thank the patients and their families for participating in the clinical studies from which the data used in these analyses were obtained. From Genentech, we thank Roel Funke and Colombe Chappey for their contribution to the development of immune-modified Response Evaluation Criteria In Solid Tumors. Colombe Chappey is currently at Pfizer (San Francisco, CA). Third-party medical writing support was provided by Meghal Gandhi (Health Interactions, San Francisco, CA), with funding from F. Hoffmann-La Roche.

1. Eisenhauer EA, Therasse P, Bogaerts J, et al: New response evaluation criteria in solid tumours: Revised RECIST guideline (version 1.1). Eur J Cancer 45:228-247, 2009 Crossref, MedlineGoogle Scholar
2. Pardoll DM: The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 12:252-264, 2012 Crossref, MedlineGoogle Scholar
3. Balar AV, Weber JS: PD-1 and PD-L1 antibodies in cancer: Current status and future directions. Cancer Immunol Immunother 66:551-564, 2017 Crossref, MedlineGoogle Scholar
4. Di Giacomo AM, Danielli R, Guidoboni M, et al: Therapeutic efficacy of ipilimumab, an anti-CTLA-4 monoclonal antibody, in patients with metastatic melanoma unresponsive to prior systemic treatments: Clinical and immunological evidence from three patient cases. Cancer Immunol Immunother 58:1297-1306, 2009 Crossref, MedlineGoogle Scholar
5. Hoos A, Eggermont AM, Janetzki S, et al: Improved endpoints for cancer immunotherapy trials. J Natl Cancer Inst 102:1388-1397, 2010 Crossref, MedlineGoogle Scholar
6. Wolchok JD, Hoos A, O’Day S, et al: Guidelines for the evaluation of immune therapy activity in solid tumors: Immune-related response criteria. Clin Cancer Res 15:7412-7420, 2009 Crossref, MedlineGoogle Scholar
7. Hodi FS, Hwu WJ, Kefford R, et al: Evaluation of immune-related response criteria and RECIST v1.1 in patients with advanced melanoma treated with pembrolizumab. J Clin Oncol 34:1510-1517, 2016 LinkGoogle Scholar
8. Nishino M, Giobbie-Hurder A, Gargano M, et al: Developing a common language for tumor response to immunotherapy: Immune-related response criteria using unidimensional measurements. Clin Cancer Res 19:3936-3943, 2013 Crossref, MedlineGoogle Scholar
9. Chen DS, Mellman I: Elements of cancer immunity and the cancer-immune set point. Nature 541:321-330, 2017 Crossref, MedlineGoogle Scholar
10. Hodi FS, O’Day SJ, McDermott DF, et al: Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med 363:711-723, 2010 Crossref, MedlineGoogle Scholar
11. Herbst RS, Soria JC, Kowanetz M, et al: Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature 515:563-567, 2014 Crossref, MedlineGoogle Scholar
12. Brahmer J, Reckamp KL, Baas P, et al: Nivolumab versus docetaxel in advanced squamous-cell non-small-cell lung cancer. N Engl J Med 373:123-135, 2015 Crossref, MedlineGoogle Scholar
13. Borghaei H, Paz-Ares L, Horn L, et al: Nivolumab versus docetaxel in advanced nonsquamous non-small-cell lung cancer. N Engl J Med 373:1627-1639, 2015 Crossref, MedlineGoogle Scholar
14. Fehrenbacher L, Spira A, Ballinger M, et al: Atezolizumab versus docetaxel for patients with previously treated non-small-cell lung cancer (POPLAR): A multicentre, open-label, phase 2 randomised controlled trial. Lancet 387:1837-1846, 2016 Crossref, MedlineGoogle Scholar
15. Rittmeyer A, Barlesi F, Waterkamp D, et al: Atezolizumab versus docetaxel in patients with previously treated non-small-cell lung cancer (OAK): A phase 3, open-label, multicentre randomised controlled trial. Lancet 389:255-265, 2017 Crossref, MedlineGoogle Scholar
16. Hoos A, Wolchok JD, Humphrey RW, et al: CCR 20th anniversary commentary: Immune-related response criteria–capturing clinical activity in immuno-oncology. Clin Cancer Res 21:4989-4991, 2015 Crossref, MedlineGoogle Scholar
17. Nishino M, Giobbie-Hurder A, Bailey N, et al: Immune-related tumor response dynamics as a marker for survival and treatment benefit during PD-1 inhibitor therapy. J Clin Oncol 34, 2016 (suppl; abstr 9521) Google Scholar
18. Seymour L, Bogaerts J, Perrone A, et al:iRECIST: Guidelines for response criteria for use in trials testing immunotherapeutics. Lancet Oncol 18:e143-e152, 2017 Crossref, MedlineGoogle Scholar
19. Tecentriq (atezolizumab) [package Insert]. South San Francisco, CA: Genentech, 2016 Google Scholar
20. Peters S, Gettinger S, Johnson ML, et al: Phase II trial of atezolizumab as first-line or subsequent therapy for patients with programmed death-ligand 1-selected advanced non-small-cell lung cancer (BIRCH). J Clin Oncol 35:2781-2789, 2017 LinkGoogle Scholar
21. Rosenberg JE, Hoffman-Censits J, Powles T, et al: Atezolizumab in patients with locally advanced and metastatic urothelial carcinoma who have progressed following treatment with platinum-based chemotherapy: A single-arm, multicentre, phase 2 trial. Lancet 387:1909-1920, 2016 Crossref, MedlineGoogle Scholar
22. Powles T, Eder JP, Fine GD, et al: MPDL3280A (anti-PD-L1) treatment leads to clinical activity in metastatic bladder cancer. Nature 515:558-562, 2014 Crossref, MedlineGoogle Scholar
23. Mazieres J, Fehrenbacher L, Rittmeyer A, et al: Non-classical response measured by immune-modified RECIST and post-progression treatment effects of atezolizumab in 2L/3L NSCLC: Results from the randomized phase II study POPLAR. J Clin Oncol 34, 2016 (suppl; abstr 9032) LinkGoogle Scholar
24. Giobbie-Hurder A, Gelber RD, Regan MM: Challenges of guarantee-time bias. J Clin Oncol 31:2963-2969, 2013 LinkGoogle Scholar
25. Nishino M, Giobbie-Hurder A, Ramaiya NH, et al: Response assessment in metastatic melanoma treated with ipilimumab and bevacizumab: CT tumor size and density as markers for response and outcome. J Immunother Cancer 2:40, 2014 Crossref, MedlineGoogle Scholar
26. Nishino M, Tirumani SH, Ramaiya NH, et al: Cancer immunotherapy and immune-related response assessment: The role of radiologists in the new arena of cancer treatment. Eur J Radiol 84:1259-1268, 2015 Crossref, MedlineGoogle Scholar
27. Erasmus JJ, Gladish GW, Broemeling L, et al: Interobserver and intraobserver variability in measurement of non-small-cell carcinoma lung lesions: Implications for assessment of tumor response. J Clin Oncol 21:2574-2582, 2003 LinkGoogle Scholar
28. Nishino M, Guo M, Jackman DM, et al: CT tumor volume measurement in advanced non-small-cell lung cancer: Performance characteristics of an emerging clinical tool. Acad Radiol 18:54-62, 2011 Crossref, MedlineGoogle Scholar
29. Oxnard GR, Zhao B, Sima CS, et al: Variability of lung tumor measurements on repeat computed tomography scans taken within 15 minutes. J Clin Oncol 29:3114-3119, 2011 LinkGoogle Scholar
30. Zhao B, James LP, Moskowitz CS, et al: Evaluating variability in tumor measurements from same-day repeat CT scans of patients with non-small cell lung cancer. Radiology 252:263-272, 2009 Crossref, MedlineGoogle Scholar
31. Hamid O, Robert C, Daud A, et al: Safety and tumor responses with lambrolizumab (anti-PD-1) in melanoma. N Engl J Med 369:134-144, 2013 Crossref, MedlineGoogle Scholar
32. Robert C, Ribas A, Wolchok JD, et al: Anti-programmed-death-receptor-1 treatment with pembrolizumab in ipilimumab-refractory advanced melanoma: A randomised dose-comparison cohort of a phase 1 trial. Lancet 384:1109-1117, 2014 Crossref, MedlineGoogle Scholar
33. Motzer RJ, Rini BI, McDermott DF, et al: Nivolumab for metastatic renal cell carcinoma: Results of a randomized phase II trial. J Clin Oncol 33:1430-1437, 2015 LinkGoogle Scholar
34. Nishino M, Ramaiya NH, Chambers ES, et al: Immune-related response assessment during PD-1 inhibitor therapy in advanced non-small-cell lung cancer patients. J Immunother Cancer 4:84, 2016 Crossref, MedlineGoogle Scholar
35. Champiat S, Dercle L, Ammari S, et al: Hyperprogressive disease is a new pattern of progression in cancer patients treated by anti-PD-1/PD-L1. Clin Cancer Res 23:1920-1928, 2017 Crossref, MedlineGoogle Scholar
36. Kato S, Goodman AM, Walavalkar V, et al: Hyperprogressors after immunotherapy: Analysis of genomic alterations associated with accelerated growth rate. Clin Cancer Res 23:4242-4250, 2017 Crossref, MedlineGoogle Scholar
Downloaded 19,171 times


DOI: 10.1200/JCO.2017.75.1644 Journal of Clinical Oncology 36, no. 9 (March 20, 2018) 850-858.

Published online January 17, 2018.

PMID: 29341833

ASCO Career Center