Article Text
PDF
Abstract
Objective To evaluate the feasibility and outcomes of performing procedural interventions, defined as surgical resection, tumor ablation, or targeted radiation therapy, for oligoprogressive disease among patients with gynecologic malignancies who are treated with immune checkpoint blockade.
Methods Patients with gynecologic cancers treated with immune checkpoint blockade between January 2013 and October 2021 who underwent procedural interventions including surgical resection, interventional radiology ablation, or radiation therapy for oligoprogressive disease were identified. Procedures performed before immune checkpoint therapy initiation or ≥6 months after therapy completion were excluded. Long immunotherapy duration prior to intervention was defined as ≥6 months. Progression-free survival and overall survival were calculated from procedure date until disease progression or death, respectively.
Results During the study period, 886 patients met inclusion criteria and received immune checkpoint blockade therapy. Of these, 34 patients underwent procedural interventions for oligoprogressive disease; 7 underwent surgical resection, 3 underwent interventional radiology ablation, and 24 underwent radiation therapy interventions. Primary disease sites included uterus (71%), ovary (24%), and cervix (6%). Sites of oligoprogression included abdomen/pelvis (26%), bone (21%), lung (18%), distant lymph node (18%), brain (9%), liver (6%), and vagina (3%). Most tumors (76%) did not exhibit microsatellite instability or mismatch repair deficiency. Approximately half (53%) of the patients had long immune checkpoint therapy duration prior to intervention. Median progression-free survival following the procedure was 5.3 months (95% CI, 3.1–9.9), and median overall survival was 21.7 months (95% CI, 14.9–not estimable). Long versus short immune checkpoint therapy duration prior to procedure and length of immune checkpoint therapy had no effect on progression-free or overall survival.
Conclusions Procedural interventions for patients with oligoprogression on immune checkpoint blockade therapy are feasible and demonstrate favorable outcomes. With expanding use of immune checkpoint therapy, it is important to investigate combined modalities to maximize therapeutic benefit for patients with gynecologic cancers.
- Gynecologic Surgical Procedures
- Radiotherapy
Data availability statement
Data are available upon reasonable request. The data that support the findings of this study are available upon reasonable request from the corresponding author (Ginger Gardner).
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Treatment of a variety of solid tumors with immune checkpoint inhibitors (programmed death-ligand 1/programmed cell death protein-1 inhibitors) has been compared with standard chemotherapy. Limited data exist on the management of patients with gynecologic malignancies who were treated with immune checkpoint blockade and then experienced oligoprogression of disease.
WHAT THIS STUDY ADDS
This is one of the first studies to report on outcomes of patients diagnosed with gynecologic cancer who received immune checkpoint blockade therapy and developed oligoprogressive disease and were subsequently treated with a variety of procedural interventions (surgical removal, directed radiation therapy, or tumor ablation).
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Procedural interventions for patients with oligoprogression during or after treatment with immune checkpoint blockade are feasible and demonstrate favorable outcomes. Continued investigation of combined systemic and tumor-directed modalities to maximize therapeutic benefit for patients with gynecologic cancers is essential.
Introduction
Immune checkpoint blockade therapy has shifted the treatment paradigm for solid malignancies over the past two decades and is now being used with increasing frequency in gynecologic cancers, particularly cervical and endometrial cancers. The United States Food and Drug Administration approved the use of pembrolizumab, a programmed cell death protein-1 (PD-1) inhibitor, in the treatment of microsatellite instability-high or mismatch repair-deficient solid tumors based on findings from the KEYNOTE-158 study, which has expanded its indication to include several gynecologic cancers.1–4 More recently, the NRG-GY018 and RUBY clinical trials demonstrated a favorable progression-free survival benefit with the addition of immune checkpoint therapy to frontline chemotherapy for metastatic or recurrent endometrial cancer.5 6
Treatment with immune checkpoint blockade may result in disease stabilization or even regression for prolonged periods compared with standard chemotherapy.7 8 Over time, however, acquired resistance at some tumor sites may occur. The best method of treatment for oligoprogressive escape lesions, defined as anatomically restricted tumor progression in patients with otherwise controlled disease with systemic therapy, remains unknown.9–11 Following the theory that oligoprogression represents disease with acquired resistance, treatment of these lesions presents an opportunity to eradicate resistant clones using local therapies.
Studies in other tumor types, such as non-small cell lung cancer and melanoma, have demonstrated that local therapy may be useful for oligoprogressive escape lesions,10 12–15 but the efficacy of surgical and procedural interventions on escape lesions after immune checkpoint therapy in gynecologic malignancies has not undergone rigorous study. Our objective was to evaluate the feasibility and outcomes of procedural interventions for oligoprogressive disease among patients with gynecologic cancer treated with immune checkpoint blockade.
Methods
We identified all patients diagnosed with cancers of Müllerian origin between January 1, 2013, and October 1, 2021, who were treated with immune checkpoint blockade, defined as a PD-1 or programmed death-ligand 1 (PD-L1) inhibitor (pembrolizumab, nivolumab, dostarlimab, cemiplimab, atezolizumab, avelumab, durvalumab, ipilimumab, tremelimumab). Patients with melanoma histology or lack of definitive disease site were excluded. All surgical, interventional radiology, or radiation therapy procedures that patients underwent between immune checkpoint therapy initiation and 6 months after treatment completion were queried. Only procedures that were performed for treatment of oligoprogressive disease were included.
Oligoprogression was defined as no more than two sites of disease growth in a background of known metastatic disease, whether by development of new sites of disease or growth of an existing site. Procedures performed before initiation of or ≥6 months after completion of immune checkpoint therapy were excluded. Patients who were treated as part of an unpublished clinical trial were also excluded.
Clinico-demographic data, treatment characteristics, and recurrence patterns were extracted from the electronic medical record and described using descriptive statistics. Microsatellite instability-high status was determined using the tumor-normal sequencing platform MSK-IMPACT (Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets).16 Microsatellite instability-high tumors were classified as having an MSIsensor score ≥10, as previously defined.17 18 For tumors that did not undergo MSK-IMPACT, mismatch repair protein immunohistochemistry was used to determine if the tumor was mismatch repair-deficient.19 Location of oligoprogressive metastasis was divided into local metastasis (abdomen, pelvis, or vagina) or distant metastasis (lung, bone, brain, liver, or distant lymph node). If patients had multiple qualifying procedures, they were assigned to the treatment group corresponding to the first procedure received, and the first procedure was used in the statistical analyses.
Survival endpoints were calculated based on the date of the first procedure. Lines of therapy prior to immune checkpoint therapy and subsequent lines of therapy were also recorded. Patients were divided based on immune checkpoint blockade therapy duration prior to procedure (long vs short). Long duration was defined as ≥6 months and short duration was defined as <6 months of immune checkpoint therapy prior to procedure. Progression-free survival was calculated from the procedure date to the date of progression, last follow-up, or death. Overall survival was calculated from the procedure date until date of death or last follow-up. Kaplan–Meier methodology was applied to obtain median progression-free survival, overall survival, and survival rate. P values for survival outcome were obtained using the log-rank test for categorical variables and the Wald test based on the Cox proportional hazards model for continuous variables. For variables with very small event numbers in a certain level, the permutation log-rank test was used.20 Due to the time-dependent nature of total immune checkpoint treatment length, the time-dependent Cox proportional hazards model was applied. A P value <0.05 was considered statistically significant. All tests were two-sided. All analyses were performed on R (version 4.2.2).21 This study was approved by the Institutional Review Board of Memorial Sloan Kettering Cancer Center (#16–1317).
In accordance with the Journal’s guidelines, the authors will provide their data for independent analysis by a selected team by the Editorial Team for the purposes of additional data analysis or for the reproducibility of this study in other centers if such is requested.
Results
We identified 1053 patients with suspected gynecologic cancer who were treated with immune checkpoint blockade. After exclusion of patients with melanoma histology (n=23) and patients with unconfirmed disease site or histology (n=144), 886 patients remained (Figure 1). Of these, 34 patients underwent procedural interventions for oligoprogressive disease, including surgery (21%, n=7), radiation therapy (71%, n=24), and tumor ablations with interventional radiology (9%, n=3) (Table 1, Figure 1). Among the 34 patients, two patients underwent two surgeries each for oligometastases, one patient had four separate radiation treatments, one patient had two separate radiation treatments, and one patient first underwent radiation followed by surgery for a separate lesion and was grouped as part of the radiation cohort.
Patient flowchart. d, days; GYN, gynecologic; ICB, immune checkpoint blockade; IR, interventional radiology ablation; RT, radiation therapy.
Patient characteristics
Median age at diagnosis was 62 years (range, 33–84). Self-reported race included White (74%, n=25), Black (12%, n=4), Asian (9%, n=3), Other (3%, n=1), and Unknown (3%, n=1). Self-reported ethnicity included non-Hispanic (91%, n=31) and Hispanic (9%, n=3) (Table 1). Primary disease site included uterus (71%, n=24), ovary (24%, n=8), and cervix (6%, n=2) (Table 1). Microsatellite instability-high/mismatch repair-deficient status was available for 33 patients (97%); 8 tumors (24%) were microsatellite instability-high/mismatch repair-deficient and 25 (76%) were not. Immune checkpoint therapy regimens included a PD-1/PD-L1 inhibitor (35%, n=12), PD-1/PD-L1 inhibitor+tyrosine kinase inhibitor (35%, n=12), PD-1/PD-L1 inhibitor+CTLA-4 inhibitor (15%, n=5), and PD-1/PD-L1 inhibitor+other (15%, n=5). Ten patients (29%) had local sites of oligoprogression, which included the abdomen or pelvis (26%, n=9) and vagina (3%, n=1). Twenty-four patients (71%) had distant sites of oligoprogression, including bone (21%, n=7), lung (18%, n=6), distant lymph node (18%, n=6), brain (9%, n=3), and liver (6%, n=2). The number of prior systemic therapy lines included 0 (12%, n=4), 1 (47%, n=16), 2 (23%, n=8), 3 (15%, n=5), and 4 (3%, n=1).
Of the 10 total surgical procedures, six were abdominopelvic debulking procedures, one was a bronchoscopy with excision of a lung mass, one a resection of a supraclavicular mass, one an inguinal lymphadenectomy, and one a craniotomy for resection of a brain metastasis. For the seven first surgical procedures in the statistical analyses cohort, surgery was performed a median of 41.6 months from diagnosis (range, 24.1–103.6), and a median of 4.6 months since the start of treatment with immune checkpoint blockade (range, 0.2–13.9) (Online supplemental table 1). Immune checkpoint therapy was stopped an average of 1.5 months prior to the procedure. All patients who underwent surgery had complete gross resection of disease, and no patients had grade 3 or higher 30-day surgical complications. After surgical excision, three patients (30%) continued the same immunotherapy regimen, three (30%) switched to chemotherapy, two (20%) switched to another immunotherapy regimen, and two (20%) underwent surveillance. In the five patients who continued with immune checkpoint therapy, treatment was restarted an average of 1.1 months after the procedure.
Supplemental material
There were 28 tumor-directed radiation therapy procedures in 24 patients. For the 24 first procedures in the statistical analyses cohort, radiation therapy was performed a median of 27.6 months after diagnosis (range, 11.0–183.7) and occurred a median of 6.6 months after the start of immune checkpoint therapy (range, 1.5–18.6). Time from start of radiation therapy to best radiographic response was an average of 77.9 days. Most patients continued the same immunotherapy regimen (54%, n=15), whereas eight patients (29%) switched to chemotherapy, three patients (11%) started other therapy, and two patients (7%) switched to another immunotherapy regimen. In patients who continued with immune checkpoint therapy, treatment was restarted an average of 1.4 months after radiation therapy.
All three interventional radiology procedures were tumor ablation and performed for oligoprogressive uterine cancer located in the lung (n=1) or liver (n=2). All three patients were treated with a PD-1/PD-L1 inhibitor in combination with a tyrosine kinase inhibitor. Median time from diagnosis to procedure was 32.9 months (range, 27.7–57.5), and median time from the start of immune checkpoint therapy to procedure was 8.9 months (range, 3.0–14.3) (Online supplemental table 1). Immune checkpoint therapy was stopped an average of 1.1 months prior to ablation. Time from ablation to best radiographic treatment response was an average of 26 days. After ablation, two patients (67%) continued the same immunotherapy regimen, which was restarted approximately 10 days after the procedure, whereas one patient (33%) went on to receive chemotherapy.
Sixteen patients (47%) had <6 months of immune checkpoint therapy prior to procedure, whereas 18 patients (53%) had a long immunotherapy duration. Median number of months between start of immune checkpoint therapy to procedure was 6.4 months (range, 0.2–18.6) for the entire statistical analyses cohort, and the median total length of immune checkpoint therapy was 7.9 months (range, 1.4–34.5); 59% of patients had an immunotherapy duration between 6 and 12 months (Table 1).
For the statistical analyses cohort, 29 progression events and one death occurred before progression, with a median progression-free survival from time of procedural intervention to next progression of 5.3 months (95% CI, 3.1–9.9). Total duration of response, calculated as time from the start of immune checkpoint therapy to progression after procedural intervention, was 15.2 months (95% CI, 10.1–19.9). Median follow-up for the four progression-free survivors was 16.7 months (range, 9.3–36.9). There were 19 deaths, with a median overall survival from time of procedural intervention to death of 21.7 months (95% CI, 14.9–not estimable). Overall survival from start of immune checkpoint blockade therapy to death was 29.8 months (95% CI, 21.9–not estimable). Median follow-up for the 15 survivors was 22.8 months (range, 5.5–65).
On univariate analyses, treatment type, age at diagnosis or immunotherapy start, self-reported race or ethnicity, disease site, metastatic site, microsatellite instability-high/mismatch repair-deficient status, long versus short immunotherapy duration prior to procedure, and length of immunotherapy had no effect on progression-free survival (Table 2). Local metastatic site and microsatellite instability-high/mismatch repair-deficient status were associated with prolonged overall survival; however, due to the small event rates, the significance should be interpreted cautiously (Table 3). Median progression-free survival for short immune checkpoint therapy duration was 6.7 months (95% CI, 2.6–12.6), compared with 4.9 months (95% CI, 2.4–9.9) for long duration (Figure 2A). Median overall survival for short immune checkpoint therapy duration was 23.0 months (95% CI, 14.3–not estimable), compared with 21.7 months for long duration (95% CI, 9.7–not estimable) (Figure 2B).
Univariate logistic regression for progression-free survival in the whole cohort
Univariate logistic regression for overall survival in the whole cohort
Kaplan–Meier curves. (A) Progression-free survival, by length of immune checkpoint blockade (ICB) treatment prior to procedure. (B) Overall survival, by length of ICB treatment prior to procedure.
Discussion
Summary of Main Results
Immune checkpoint blockade therapy has changed the paradigm of gynecologic cancer treatment. Although several patients treated with immune checkpoint inhibitors demonstrate durable responses, the best treatment for oligoprogressive disease sites, which are thought to harbor treatment-resistant tumor clones, is not known. In this study, we describe treatment patterns for a heterogeneous cohort of 34 patients with non-melanoma gynecologic malignancies treated with immune checkpoint blockade who had oligoprogressive metastases. Most tumors were not microsatellite instability high or mismatch repair-deficient. Our findings demonstrate that procedural interventions consisting of surgery, radiation therapy, or tumor ablation are feasible, safe, and result in durable effects on progression-free survival as well as overall survival.
Results in the Context of Published Literature
To our knowledge, this is one of the first studies to examine procedural interventions in immune checkpoint inhibitor-treated gynecologic malignancies. Immune checkpoint blockade therapy has long been used in the treatment of other solid tumor types, and the benefits of local therapy on oligoprogressive lesions have been demonstrated in patients diagnosed initially with melanoma, lung, liver, and prostate cancers.14 22–24 In a cohort of patients with stage IV non-small cell lung cancer who developed oligoprogressive lesions on immune checkpoint therapy, treatment with continued immunotherapy and local radiation therapy demonstrated prolonged progression-free survival (15.0 vs 10.3 months) and overall survival (26.4 vs 20.8 months) compared with other treatment strategies.12 In a separate study of oligoprogressive lesions in a heterogeneous cohort of solid tumors treated with immune checkpoint blockade, ablative radiation therapy was found to be feasible and resulted in a median progression-free survival of 7.1 months and a 1-year overall survival rate >96%.25 Most patients in the study continued immune checkpoint therapy after radiation therapy, and at second progression, 15 of 17 patients did so at three or fewer metastatic sites and could have benefitted from additional salvage radiation therapy to further extend the lifespan of immune checkpoint blockade treatment.
In addition to radiation therapy, other local therapies including surgery, Y90 radioembolization, transarterial chemoembolization, radiofrequency ablation, and cryoablation have also been described with success in oligoprogressive metastases for a variety of tumor types initially treated with immune checkpoint blockade.26 Similarly, our findings suggest that the procedural interventions performed for oligoprogressive lesions are safe and feasible; no grade 3 complications requiring further procedures or re-operation were observed, complete gross resection was achieved at all surgical cytoreductive procedures, and excellent tumor response to interventional radiology and radiation therapy treatments was observed on subsequent imaging scans.
The length of time between start of immune checkpoint therapy to procedure may be considered as a surrogate for disease aggressiveness and therapy effectiveness. The median time from start of immune checkpoint blockade therapy to procedure in our cohort was 6.4 months. Our analyses demonstrated no difference between short or long immunotherapy duration before procedure. Although systemic therapy treatments varied after the procedure, most patients had a total immune checkpoint therapy length between 6 and 12 months. Our results are similar to published data demonstrating that time to oligoprogression is not a significant predictor of progression-free survival after local therapy,25 suggesting that procedural interventions may benefit even patients with biologically more aggressive disease.
Although the number of patients in our cohort is small, on exploratory analyses investigating factors associated with prolonged progression-free survival and overall survival, we found that microsatellite instability-high/mismatch repair-deficient status was associated with prolonged overall survival. Similarly, having a local metastatic site was associated with prolonged overall survival, though the number of events is small, and the data must be interpreted cautiously.
Strengths and Weaknesses
Strengths of this study include starting with a large and diverse cohort of patients with gynecologic cancers who were treated with immune checkpoint blockade therapy. Our cohort was generated after clinician review of all procedures performed, and only those performed for oligoprogression were included. To our knowledge, this report is the first to study the impact of procedural interventions for oligoprogression in patients with gynecologic-specific cancers who are treated with immune checkpoint therapy. Limitations of this study include the use of a small and heterogeneous cohort of patients with a variety of primary tumor locations, histologies, immune checkpoint inhibitor-containing regimens, and procedural interventions, which may limit the generalizability of the study findings. Patients may have received immune checkpoint therapy at any time during their treatment course, which confounds survival analyses. Our progression-free survival and overall survival results are exploratory and should be interpreted with caution. Although use of data from a single institution allows for comprehensive data capture through medical record review, generalizability may be limited, as it is a single-site study.
Implications for Practice and Future Research
Future directions of study include comparison of oligoprogression rates with immune checkpoint monotherapy as opposed to combined therapy, and more long-term follow-up of patients who receive procedural interventions for oligoprogression. Additionally, the best therapy modality following procedural intervention for oligoprogression (switching to alternate therapy vs remaining on same therapy vs surveillance) remains unknown. In addition, molecular analysis of the differential response of oligoprogressive lesions deserves further investigation. In this study, we demonstrate that among patients with gynecologic cancer, procedural interventions for oligoprogressive disease are feasible, safe, and allow patients to then continue immune checkpoint blockade therapy or other meaningful systemic therapy.
Conclusions
The findings of this early report suggest that procedural interventions for patients with gynecologic malignancies who experience oligoprogression on immune checkpoint blockade therapy are feasible and demonstrate favorable outcomes. With the expanding use of immune checkpoint therapy, it is important to investigate combined modalities to determine which combinations maximize therapeutic benefit for patients with specific gynecologic cancers. Mechanisms to identify the best candidates for procedural intervention have yet to be defined and this is a promising avenue for future research.
Data availability statement
Data are available upon reasonable request. The data that support the findings of this study are available upon reasonable request from the corresponding author (Ginger Gardner).
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by the Institutional Review Board of Memorial Sloan Kettering Cancer Center (#16-1317). Participants gave informed consent to participate in the study before taking part.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Presented at This work was presented as a Featured Poster at the 2022 International Gynecologic Cancer Symposium in New York, NY, USA.
Contributors Conceptualization: TYS, DZ, GG. Data curation: TYS, VW, OZ, YS, DSC, KL, EJ, WT, RO, SC, VM, YLL, CFF, CK, DZ, GG. Formal analysis: TYS, QCZ, AI. Methodology: TYS, DZ, GG. Supervision: DZ, GG. Writing – original draft: TYS, VW, MF. Writing – reviewing and editing: TYS, MF, QCZ, YS, DSC, EJ, RO, DZ, GG. GG is responsible for the overall content as guarantor.
Competing interests Outside the submitted work, AI reports consulting fees from Mylan. CFF reports institutional research support from Seagen, Merck, BMS, AstraZeneca, Mersana, and Hotspot Therapeutics; consulting fees from BMS, Seagen, and Aadi Biosciences; honoraria for lectures from Onclive; meeting/travel support by Puma; and participation on Data Safety Monitoring or Advisory Board of Merck, Genentech, and Marengo (all uncompensated). DZ reports institutional research support from AstraZeneca, Merck, Plexxikon Synthekine, and Genentech; consulting fees from AstraZeneca, Synthekine, Astellas, Tessa Therapeutics, Memgen, Celldex, Crown Biosciences, Hookipa Biotech, Kalivir, Xencor, and GSK; royalties from Merck; and stock options from Accurius Therapeutics, ImmunOS Therapeutics, and Calidi Biotherapeutics. VM reports meeting/travel support by Eisai and Merck; participation on Data Safety Monitoring or Advisory Board of Duality, Merck, Karyopharm, Exelexis, Eisai, Karyopharm, BMS, Clovis, Faeth Immunocore, Morphosys, AstraZeneca, Novartis, GSK, and Bayer (all unpaid); and study support to the institution by Merck, Eisai, AztraZeneca, Faeth, Karyopharm, Zymeworks, Duality, Clovis, Bayer, and Takeda. YLL reports institutional research funding from Repare Therapeutics, AstraZeneca, and GSK; honoraria from Total Health and Sarah Lawrence College; and travel/meeting support by AstraZeneca. DSC reports personal fees from Apyx Medical, Verthermia Inc., Biom ‘Up, and AstraZeneca, as well as recent or current stock/options ownership of Apyx Medical, Verthemia, Intuitive Surgical, Inc., TransEnterix, Inc., Doximity, Moderna, and BioNTech SE. EJ reports personal fees from Covidien/Medtronic. RO reports personal fees from Tesaro/GSK, Regeneron, R-PHARM, Seattle Genetics, Fresenius Kabi, Gynecologic Oncology Foundation, Bayer, Curio, Miltenyi, 2seventybio, and Immunogen; and other from Hitech Health, all outside the submitted work; non-compensated steering committee membership for the PRIMA, Moonstone (Tesaro/GSK), and DUO-O (AstraZeneca) studies; a non-compensated advisor role for Carina Biotech; and funding for clinical research from Bayer/Celgene/Juno, Tesaro/GSK, Merck, Ludwig Cancer Institute, Abbvie/StemCentrx, Regeneron, TCR2 Therapeutics, Atara Biotherapeutics, MarkerTherapeutics, Syndax Pharmaceuticals, Genmab/Seagen Therapeutics, Sellas Therapeutics, Genentech, Kite Pharma, Acrivon, Lyell Immunopharma, and Gynecologic Oncology Foundation. CK reports grant funding from Conquer Cancer Foundation; grant funding paid to the institution from Merus, Gritstone, and BMS; and consulting fees from Scenic Immunology B.V. and OncLive. All other authors have no potential conflicts of interest to disclose.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.