Immunotherapy After Immunotherapy: Response Rescue in a Microsatellite Instability-High Colorectal Cancer Patient Post-Pembrolizumab

Abstract

Background:

Metastatic microsatellite instability-high (MSI-H) colorectal cancer (CRC) is one of the rare gastrointestinal tumors where immune checkpoint inhibitors (ICIs) have a defined therapeutic role. Single-agent nivolumab, single-agent pembrolizumab and the combination of nivolumab plus ipilimumab are all FDA-approved therapies in patients with refractory disease. Limited data exists however, on how to treat patients who progress on anti-programmed cell death protein 1 (PD-1) therapy, specifically with regards to continuing immunotherapy. We present the case of a young woman with MSI-H CRC who received nivolumab plus ipilimumab post-progression on pembrolizumab and remains in partial response. To the best of our knowledge, this is the first report of such a phenomenon in a CRC patient.

Case Presentation:

A 39-year-old female with MSI-H (loss of MSH2 and MSH6 by initial immunohistochemistry) cecal adenocarcinoma started therapy with pembrolizumab in February 2018 after progressing on a clinical trial with fluorouracil and irinotecan (FOLFIRI) plus a matrix metalloproteinase-9 inhibitor. In October 2018, although she met criteria for stable disease by RECIST 1.1 on CT scans, she began to exhibit mild growth in her lung metastases. Because she had been tolerating pembrolizumab well, she was continued on the therapy. In December 2018 she developed overt disease progression in her liver and was taken off the anti-PD-1 antibody. Given her tumor mutational burden (55 Muts/Mb) and absence of other meaningful options, her therapy was switched to nivolumab plus ipilimumab (240 mg and 1 mg/kg every 3 weeks, respectively). She received 4 doses of the combination and her first restaging scans in March 2019 demonstrated stable disease (20% tumor reduction). She was transitioned to single agent nivolumab every 2 weeks thereafter and demonstrated a partial response (30% tumor reduction) on her subsequent CT scans in June 2019.

Conclusion:

We have found no prior published reports of MSI-H CRC patients responding to combination immunotherapy with nivolumab plus ipilimumab after prior progression on anti-PD-1 antibodies. This approach has been successful in metastatic melanoma patients and warrants prospective clinical trial evaluation in MSI-H CRC patients to assess whether it can influence future post-ICI progression treatment strategies.

Introduction:

Nivolumab, pembrolizumab and nivolumab plus ipilimumab are all FDA-approved options for the treatment of refractory metastatic microsatellite instability-high (MSI-H) colorectal cancer (CRC). Response rates in patients treated with single-agent anti-programmed cell death protein 1 (PD-1) antibodies are 31–32%; response rate in patients treated with the combination is 55% [1,2,3]. Despite this difference, many providers start with single-agent therapy due to lower rates of grade 3/4 adverse events and similar 12-month OS rates. Responding patients often experience durable responses however most patients develop progressive disease. How to proceed after patients progress on anti-PD-1 antibodies remains undefined. Specifically, the question of whether combinatorial immune checkpoint inhibitor (ICI) therapy can rescue patients after progression on single-agent anti-PD-1 antibodies is especially salient in MSI-H CRC where limited later-line treatment options promising sustained benefit are available.

Insights from metastatic melanoma (mel) patients suggest that patients with progressive disease after treatment with anti-PD-1 antibodies can be salvaged with the combination of anti-cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) plus anti-PD-1 antibodies [4,5]. In a recent prospective phase II single arm study, 45% of mel patients who progressed on anti-PD-1 first-line therapy achieved responses by RECIST 1.1 after treatment with pembrolizumab 200 mg every 3 weeks and ipilimumab 1 mg/kg every 3 weeks (for 4 doses). Median progression-free survival (PFS) was 7.2 months in the cohort while median overall survival (OS) was not reached. In other diseases, such as renal cell carcinoma [ NCT03117309, NCT03177239] and non-small cell lung cancer [ NCT04151563], the strategy of treating patients who have progressed on single-agent anti-PD-1 antibody therapy with dual anti-CTLA-4 and anti-PD-1 therapy is also being trialed. To the best of our knowledge, there are no reports of utilizing this strategy in MSI-H CRC patients who have progressed on single-agent anti-PD-1 antibodies. We report the case of a young woman who is benefiting from nivolumab plus ipilimumab post-progression on pembrolizumab.

Case Description:

A 39-year-old woman with MSI-H (loss of MSH2 and MSH6 by immunohistochemistry) liver-limited metastatic disease received curative-intent therapy with peri-operative fluorouracil and oxaliplatin (FOLFOX) for 6 cycles prior to undergoing resection of her liver metastases on August 4, 2016. She completed 6 further cycles of FOLFOX on December 8, 2016. Unfortunately, she was noted to have progressive disease in her liver and lungs on restaging CT scans done shortly thereafter on December 18, 2016. Next-generation sequencing was performed on one of her liver metastectomy specimens via the Foundation One assay. The sequencing results redemonstrated MSI-H status, a tumor mutational burden (TMB) of 55 Muts/Mb and the presence of a BRCA1 mutation. She subsequently initiated a clinical trial with fluorouracil and irinotecan (FOLFIRI) plus a matrix metalloproteinase-9 inhibitor and remained on the study until January 10, 2018 when, in the setting of difficulty tolerating the therapy and mildly progressive pulmonary lesions, she began treatment with single agent pembrolizumab 200 mg every 3 weeks. She tolerated pembrolizumab well, except for grade 2 arthralgias that required two short steroid courses (5-days of prednisone 20 mg) in June 2018 and October 2018. The patient achieved stable disease on her CT scans from October 10, 2018 however began to demonstrate mild growth of her pulmonary nodules. She developed overt disease progression in her liver on CT scans from December 29, 2018. In the absence of other options which could offer her sustained disease control, initial period of disease stability while on anti-PD-1 therapy and robust performance status, she was started on combination immunotherapy with nivolumab plus ipilimumab. The patient received nivolumab 240 mg every 3 weeks along with ipilimumab 1 mg/kg every 3 weeks for 4 doses prior to transitioning to nivolumab maintenance every 2 weeks. The patient’s first restaging CT scans on March 23, 2019, after she completed the 4 doses of combination therapy, demonstrated stable disease (20% tumor reduction) with shrinkage of her dominant liver and left ovarian metastases (Figure 1 A–F). Her subsequent scans on June 1, 2019 after 4 doses of nivolumab monotherapy revealed a partial response (30% tumor reduction) (Figure 2 A–C).

Figure 1:

Abdominal and pelvic CT images of patient’s dominant liver lesions and left ovarian lesion. In panels A, C and E are images at time of overt progression on pembrolizumab. In panels B, D and F are images of the same lesions after 4 treatments of nivolumab and ipilimumab. Measurement bars indicate the size of each lesion in centimeters.

Figure 2:

The most recent abdominal and pelvic CT images of the patient’s dominant liver lesions and left ovarian lesion after 4 treatments of maintenance nivolumab. In panels A and B are images of the patient’s dominant liver lesions. In panel C is an image of the patient’s left ovarian lesion. Measurement bars indicate the size of each lesion in centimeters.

Discussion:

Our report highlights the case of a woman with MSI-H CRC who responded to nivolumab plus ipilimumab after previously progressing on pembrolizumab. Although we recognize the limitations of overextending conclusions from a single patient case, no prior cases of this phenomenon have been documented in MSI-H CRC patients. Furthermore, there appears to be precedent for combining anti-CTLA-4 antibodies with anti-PD-1 antibodies to overcome acquired anti-PD-1 antibody tumor resistance based on the experience from mel patients [4,5,6]. The narrative for exploring this treatment strategy in MSI-H CRC patients may only become more compelling with pending studies, utilizing the same approach, in renal cell carcinoma and non-small cell lung cancer patients.

The case raises three primary issues. The first pertains to the importance of distinguishing between primary and acquired anti-PD-1 antibody tumor resistance. Tumors develop each mode of resistance through disparate mechanisms, which potentially carry implications for how to treat patients in subsequent treatment lines [7,8,9]. In primary ICI resistance, tumors are unrecognized by infiltrating T-cells due to limited antigen exposure or masking of antigen presentation. In acquired ICI resistance, tumors escape immune surveillance through intrinsic and extrinsic pathways; tumor intrinsic mechanisms include upregulation of mitogen-activated protein kinase and WNT pathway signaling while tumor extrinsic mechanisms include increased infiltration of immunosuppressive cells (T-regulatory cells, tumor associated macrophages and myeloid derived suppressor cells), induction of fixed epigenetic states in T-cells that prevent reactivation and upregulation of T-cell inhibitory signals such as CTLA-4, lymphocyte-activation gene 3 (LAG-3) and T-cell immunoglobulin and mucin-domain containing-3 (TIM-3) [10,11]. These differential mechanisms perhaps explain why adding anti-CTLA-4 antibodies to anti-PD-1 antibodies in certain patients can overcome acquired but not primary anti-PD-1 antibody resistance [12,13,14]. The patient in the case achieved stable disease on pembrolizumab for 8 months prior to progression, suggesting that her tumor developed acquired resistance to the agent. We do not know the specific mechanism by which her tumor developed anti-PD-1 antibody resistance given that a tissue biopsy was not obtained at the time of progression.

The second issue pertains to why the described patient’s disease progressed on single-agent ICI therapy initially. Her TMB of 55 Muts/Mb exceeds the mutational burden that is thought to be predictive of single-agent response to anti-PD-1 therapy in MSI-H CRC. In a retrospective analysis of 22 MSI-H CRC patients treated with ICIs, a TMB threshold of 37.4 Muts/Mb was identified as a potential cutoff for predicting benefit from anti-PD-1 antibody therapy [15]. In this analysis, among the 59% of patients with tumors with a TMB > 37.4 Muts/Mb, all patients achieved objective responses with anti-PD-1 antibody therapy; median progression-free survival after 18 months of follow-up had not yet been reached. One possible reason for her disease progression may have been due to inter-tumor TMB heterogeneity. Though the tumor that was sampled for TMB assessment through next-generation sequencing may have expressed an extraordinarily high neoantigen burden, this may not have necessarily been the case for a tumor from a different metastatic site, which would enable disparate responses to anti-PD-1 therapy within the same patient. A recent analysis suggests inter-tumor TMB heterogeneity may be present in up to 17% of cases, making it difficult to generalize about the overall neoantigen burden, and predict ICI responsiveness, of a metastatic tumor from a single metastatic site biopsy [16].

The final issue pertains to the paucity of clinical trials in MSI-H CRC that allow patients with prior ICI exposure to enroll. The landscape-shaping Checkmate 142 study has a variety of arms such as nivolumab monotherapy, nivolumab plus ipilimumab, nivolumab plus ipilimumab in combination with other immune modulating agents, and the immunotherapy combinations in frontline versus later-line settings, but excludes patients who have previously progressed on anti-PD-1 antibodies [ NCT02060188]. This is true of most studies, leading to the exclusion of a significant number of MSI-H CRC patients; 60–70% of patients demonstrate primary resistance while 30–40% of patients demonstrate acquired resistance to anti-PD-1 antibody therapy [1,2]. We believe both these patient subgroups need to be identified and prospectively evaluated in future studies, and would recommend mandatory biopsies at time of enrollment, in order to elucidate resistance mechanisms. Once particular resistance mechanisms, or patterns of resistance, are identified, novel therapeutic strategies to overcome these mechanisms can be tested, fostering potential drug development opportunities for other checkpoint agonists, antagonists and metabolic manipulators [14].

Conclusion

Treating metastatic MSI-H CRC patients post-anti-PD-1 antibody progression remains a major challenge. This challenge will remain despite efforts from investigators to move anti-PD-1 antibodies into the frontline setting for MSI-H CRC patients [17]. Most existing trials in this population exclude patients who have received prior anti-PD-1 antibodies, which may represent a missed opportunity to learn about overcoming ICI resistance in this patient group. Malcolm Gladwell defined a tipping point as “…that magic moment when an idea…crosses a threshold, tips, and spreads like wildfire [18].” Although our report highlights the case of a single MSI-H CRC patient whose tumor anti-PD-1 antibody resistance was overcome by dual ICI blockade with the addition of an anti-CTLA-4 antibody, perhaps her case can serve as a tipping point to inspire dedicated clinical trials that examine ICI resistance mechanisms in this population.

Clinical Practice Points.

• Nivolumab, pembrolizumab and nivolumab plus ipilimumab are all FDA-approved treatment options for refractory metastatic microsatellite-instability high (MSI-H) colorectal (CRC) cancer patients.

• Most MSI-H CRC patients are treated with single-agent anti-programmed cell death 1 (PD-1) antibodies. Despite durable responses in responders, the majority of patients develop resistance. Existing guidelines offer limited input about whether immune checkpoint inhibitors (ICIs) can be continued post-anti-PD-1 antibody progression in MSI-H CRC patients and most current clinical ICI trials in this population, exclude patients who have previously received anti-PD-1 antibodies.

• Prospective data from metastatic melanoma patients suggests patients who develop acquired resistance to anti-PD-1 antibodies can be salvaged with the combination of anti-cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) plus anti-PD-1 antibodies. To the best of our knowledge, there have been no published reports of this strategy being utilized successfully in MSI-H CRC patients.

• We report the case of a young woman with metastatic MSI-H CRC who has achieved a partial response with nivolumab plus ipilimumab after progressing on pembrolizumab. Although her case represents a single patient experience, the promise of this treatment strategy warrants prospective evaluation in MSI-H CRC patients given the limited available treatment options post-progression on ICIs that offer the potential for sustained disease control.