Assessing male gamete genome integrity to ameliorate poor assisted reproductive technology clinical outcome

Abstract

Objective:

To assess the role of evaluating sperm chromatin fragmentation (SCF) as a tool to guide treatment in couples who achieved unexpectedly poor clinical outcomes after intracytoplasmic sperm injection (ICSI).

Design:

We identified couples with an unexpectedly suboptimal clinical outcome after ICSI who were then screened for SCF. Consequently, the same couples were counseled to undergo a subsequent ICSI cycle using either ejaculates processed by microfluidic sperm selection (MFSS) or spermatozoa retrieved from the testis, and clinical outcomes were compared between history and treatment cycles. To confirm the sole effect of a compromised male gamete, we compared the ICSI outcome in cycles where male gametes with abnormal SCF were used to inseminate autologous and donor oocytes. Finally, to eliminate an eventual confounding female factor component, we compared the clinical outcome of ICSI cycles using sibling donor oocytes injected with spermatozoa with normal or abnormal SCF.

Setting:

Academic reproductive medicine center point of care.

Patient(s):

The patient population consisted of 76 couples with reproductively healthy and relatively young female partners and male partners with compromised semen parameters, but suitable for ICSI. In a subanalysis, we identified 67 couples with abnormal SCF who underwent ICSI cycle(s) with donor oocytes. Furthermore, we identified 29 couples, 12 with normal SCF and 17 with abnormal, uncorrected SCF, and 7 couples with abnormal, corrected SCF vs. a control, who used sibling donor oocytes for their ICSI cycle(s).

Intervention(s):

For couples who resulted in surprisingly low clinical outcomes after ICSI, despite semen parameters adequate for ICSI and a normal female infertility evaluation, a SCF assessment was performed on the semen specimen using the terminal deoxynucleotidyl transferase-mediated fluorescein-deoxyuridine triphosphate nick-end labeling (TUNEL) assay. The couples then underwent a subsequent ICSI cycle with spermatozoa processed by MFSS or surgically retrieved. Moreover, cycles with donor oocytes were used to confirm the sole contribution of the male gamete.

Main Outcome Measure(s):

Clinical outcomes, such as fertilization, embryo implantation, clinical pregnancy, delivery, and pregnancy loss rates were compared between history and treatment cycle(s) using ejaculated spermatozoa selected by MFSS or from a testicular biopsy, taking into consideration the level of SCF. In a subanalysis, we reported the clinical outcomes of 67 patients who used donor oocytes and compared them with cycles where their own oocytes were used. Furthermore, we compared the ICSI clinical outcomes between cycles using sibling donor oocytes injected with low or high SCF with or without sperm intervention aimed at correcting, or alleviating the degree of SCF.

Result(s):

In a total of 168 cycles, 76 couples had in a prior cycle a 67.1% fertilization rate, and clinical pregnancy and pregnancy loss rates of 16.6% and 52.3%, respectively. After testing for SCF, the DNA fragmentation rate was 21.6%. This led to a subsequent ICSI cycle with MFSS or testicular sperm extraction, resulting in clinical pregnancy and delivery rates of 39.2%, and 37.3%, respectively. The embryo implantation rate increased to 23.5%, whereas the pregnancy loss rate decreased to 5% in the treatment cycle. This was particularly significant in the moderate SCF group, reaching embryo implantation, clinical pregnancy, and delivery rates of 24.3%, 40.4%, and 36.2%, respectively, and reducing the pregnancy loss rate to 10.5% in post–sperm treatment cycles.

In 67 patients with high SCF who used donor oocytes, a significantly higher fertilization rate of 78.1% and embryo implantation rate of 29.1% were reported, compared with those in couples also with an elevated SCF who used their own. Interestingly, the clinical pregnancy and delivery rates only increased slightly from 28.0%–36.1% and from 23.7%–29.2%, respectively.

To further control for a female factor, we observed couples who shared sibling donor oocytes, 17 with normal SCF and 12 with abnormal (uncorrected) SCF. Interestingly, the abnormal SCF group had impaired fertilization (69.3%), embryo implantation (15.0%), and delivery (15.4%) rates.

For an additional 15 couples who split their donor oocytes, 8 had normal SCF, and although 7 couples originally had abnormal SCF, 4 used microfluidic processing, 2 used testicular spermatozoa, and 1 used donor spermatozoa to alleviate the degree of SCF, resulting in comparable clinical outcomes with the normal SCF group.

Conclusion(s):

A superimposed male factor component may explain the disappointing ICSI outcome in some couples despite reproductively healthy female partners. Therefore, it may be useful to screen couples for SCF to guide treatment options and maximize chances of a successful pregnancy. The improved, but suboptimal pregnancy and delivery outcomes observed in couples using donor oocytes confirmed the exclusive detrimental role that the male gamete exerted on embryo development despite the presence of putative oocyte repair mechanisms.

Several advancements have been made in the field of human reproduction, particularly in relation to obviating weaknesses of the male gamete or addressing oocyte aneuploidy. Nowadays, it is feasible to treat most forms of male factor infertility, particularly in cases with a very low concentration or impaired characteristics of the spermatozoon. Indeed, even those men with severely dysmorphic gametes, such as the globozoospermic, can be successfully treated by ICSI (1), and even scarce sperm cells retrieved from the seminiferous tubules of nonobstructive azoospermic individuals can be used to successfully inseminate an oocyte (2).

However, it is most challenging to treat couples who are still unable to reproduce using assisted reproductive technology (ART), even if they present with a male partner with adequate semen parameters and a female partner of a relatively young age and negative infertility workup. These couples are initially often treated by timed intercourse or intrauterine insemination (IUI), and when those approaches fail, they are enrolled in a more aggressive treatment without a clear understanding of the reason for their reproductive failure.

Assessment of the chromatin integrity of the spermatozoa has become more demanding in recent practice. This can be performed at the first infertility evaluation, despite the fact that this approach remains debatable according to the guidelines of the American Society for Reproductive Medicine (3), or immediately after the actual first treatment attempt failure (4). SCF refers to the presence of single-stranded deoxyribonucleic acid (ssDNA) or double-stranded deoxyribonucleic acid (dsDNA) breaks in the less coiled, damage-prone linker regions bound by histones instead of by protamine. There are different bioassays available to assess SCF. The sperm chromatin structure assay is considered the gold standard to measure SCF, and the Comet assay is the most sensitive, where alkaline Comet assesses both dsDNA/ssDNA breaks and neutral Comet measures solely dsDNA breaks. However, the most used assay is TUNEL, because it is the least subjective and capable of assessing the SCF in samples with low concentrations of spermatozoa, such as in surgically retrieved testicular samples.

The observed presence and eventually the degree of SCF can then help to guide treatment approaches, such as retrieving spermatozoa directly from the testicle (5) or the selection of spermatozoa by microfluidics (6). In our previous study, testicular spermatozoa are found to have the lowest level of SCF because of less exposure to reactive oxygen species presenting in the male genital tract. Indeed, surgically retrieved spermatozoa greatly improved the clinical outcome for men with high SCF in ejaculates (5). However, surgical sperm retrieval is not always palatable considering potential surgical complications and financial burden. Therefore, the noninvasive approach using microfluidic-based technology offers an alternative for men with high SCF in their ejaculates. As previously described by Palermo et al. (7), sperm motility and SCF are inversely related. With the use of a mesh-based microfluidics system, only spermatozoa with the highest progressive motility can pass through the barrier, hence selecting spermatozoa with the lowest level of SCF. This method has been shown to greatly improve clinical outcomes for couples plagued by high SCF by yielding a higher number of euploid embryos, leading to higher implantation, clinical pregnancy, and delivery rates (6). There are indeed encouraging results with both methods. Moreover, when reproductive timing is not an issue, an even more conservative, but less consistent treatment is represented by the administration of antioxidants (8).

In this study, we investigated the role of testing for male gamete chromatin integrity to elucidate reasons for poor ART outcome and help guide the treatment approach. We identified ART cycles performed by ICSI with a poor embryo development and disappointing clinical outcome in couples with a male partner with adequate semen samples and a relatively young female partner with a negative infertility workup. We then executed a sperm chromatin integrity assay, and according to the score, we followed up with embryological and clinical outcomes. In subsequent cycles, we treated couples by collecting sperm specimens directly from the testicles or by processing the ejaculated specimen utilizing a microfluidic chamber. The resulting embryological outcome of ICSI cycles after treatment was compared with the history cycles. To determine the effect of sperm DNA fragmentation on embryo development while controlling for female gamete–induced confounding factor, in a subanalysis, we evaluated the clinical outcome of ICSI cycles performed with donor oocytes.

MATERIALS AND METHODS

Inclusion Criteria and Study Design

In our study, we included 76 couples who underwent ICSI cycles with disappointing results at the Ronald O. Perelman and Claudia Cohen Center for Reproductive Medicine from 2004–2021. The median maternal age was 35 (interquartile range [IQR], 32–36 years) years, and the median paternal age was 40 (IQR, 36–50 years) years. All female partners had a normal body mass index and infertility workup with no history of smoking, drinking, or recreational drug usage.

An initial infertility evaluation was performed on all female partners through a comprehensive review of medical, reproductive, and family history. A targeted physical examination was performed to detect pathology that could potentially affect reproductive competence. Subsequent diagnostic evaluation was conducted, including tests of ovulatory function, ovarian reserve, tubal patency, and uterine abnormalities. All female partners had a comparable duration and indication of infertility, hormonal profiles, and superovulation protocol. Male partners had a semen analysis assessed by World Health Organization (WHO) standards and had semen parameters adequate for ICSI (9). This study comprised participants of mostly White or Asian descent.

These 813 couples underwent ICSI cycle(s) where spermatozoa were selected by density gradient. On the basis of the poor clinical results obtained after their ICSI treatment, we decided to screen the male partner for SCF by TUNEL. Patients with normal SCF were excluded (n = 737). On the basis of the obtained elevated SCF results, 76 male partners were counseled and offered to undergo testicular sperm retrieval (n = 13) or MFSS (n = 63). In a subsequent cycle, according to the degree of SCF (moderate, 15%–30%, and severe, >30%), clinical outcomes were reported. The clinical outcome of all history ICSI cycles before SCF assessment, the first cycle immediately after SCF assessment, and all cycles thereafter were compared entailing fertilization, embryo implantation, clinical pregnancy, delivery, and pregnancy loss rates.

To control for an eventual confounding female factor, a subset of cycles performed with donor oocytes in men (n = 67) with compromised chromatin fragmentation were analyzed and compared with a control. To further recognize the sole contribution of the male gamete, we compared the ICSI clinical outcomes for a total of 29 couples who split donor oocytes, 17 of whom had normal SCF and 12 of whom had abnormal, uncorrected SCF. Finally, we compared the ICSI clinical outcomes of 14 patients, 8 of whom had normal SCF and 7 of whom initially had abnormal SCF but used MFSS (n = 4), testicular spermatozoa (n = 2), or donor spermatozoa (n = 1) to reduce SCF for the respective cycle.

This study was approved by the institutional review board (IRB) at Weill Cornell Medicine (IRB# 0712009553 and IRB# 1006011085). All patients were appropriately counseled and signed a consent in agreement to participate.

Spermatozoa Collection and Preparation

Fresh ejaculates were obtained by masturbation with a mean abstinence period of 2 days for all SCF groups. The specimen was then incubated in 37 ^°C for 15 minutes to allow for liquefaction. For an initial analysis on the raw semen specimen, volume, concentration, and motility were evaluated according to the WHO standards (9).

Density gradient.

Liquefied semen samples were diluted with human tubal fluid medium (Irvine Scientific, Santa Ana, CA) supplemented with human serum albumin (human serum albumin solution G Series culture media; Vitrolife, Göteborg, Sweden) and centrifuged at 600 × g for 10 minutes to remove seminal plasma. To select for motile spermatozoa, the resulting pellet was resuspended into fresh medium and loaded onto 2-layer discontinuous density gradient (Enhance-S Plus Cell Isolation Media, 45%, 90% Vitrolife) if the sample had normal or adequate semen parameters. For a severe oligozoospermic sample (<5 × 10⁶/mL), a single 90% layer was used. Once loaded onto the appropriate density gradient, the sample was centrifuged once again at 300 × g for 10 minutes. Motile spermatozoa were then isolated from the density gradient layer and resuspended in media before a final centrifugation at 600 × g for 10 minutes to remove residual silica particles. The supernatant was discarded, and the resulting pellet was resuspended in 0.5 mL of media, after which the concentration and motility were assessed on 5 μL of the final sample using a Makler chamber.

Microfluidic sperm selection.

Microfluidic sperm selection was performed as previously described (6) using a commercial device (ZyMōt Multi [850 μL] device; DxNow, Gaithersburg, MD). The semen sample was incubated for 20–30 minutes at 37 °C to allow for liquefaction. A 1 mL syringe was used to draw 850 μL of the liquefied semen sample that was then gently loaded into the device Inlet Port. Human tubal fluid medium was loaded to completely cover the top of the membrane. The loaded device was incubated for 30 minutes at 37 °C and 90% humidity. After incubation, motile spermatozoa were obtained from the harvest chamber with a syringe. The collected sample was reassessed for concentration, motility, morphology, and DNA fragmentation. In preparation for ICSI, the concentration of spermatozoa was adjusted to 1–3 × 10⁶/mL.

Testicular biopsy.

Testicular biopsy was performed as previously described (5, 10). Briefly, the seminiferous tubules were dissected, transferred into a suspension, and assessed for the presence of spermatozoa under a phase-contrast microscope at 200× magnification. Further extraction was performed from the same testis and eventually the contralateral testis if spermatozoa were not immediately detected. If no spermatozoa were identified, the testicular tissue was digested with collagenase.

SCF Assay

The terminal deoxynucleotidyl transferase-mediated fluorescein-deoxyuridine triphosphate nick-end labeling (TUNEL) assay was used to detect and quantify DNA fragmentation in the semen specimen. TUNEL was performed using a commercially available kit (In Situ Cell Death Detection Kit, Fluorescein; Roche Diagnostics, Rotkreuz, Switzerland), as previously described (6). Slides were smeared with 5–7 μL of the semen specimen with a concentration of 1 × 10⁷–2 × 10⁷ sperm/mL and dried overnight at room temperature. Next, the slides were fixed in 4% paraformaldehyde for 1 hour at room temperature and washed 3 times in phosphate-buffered saline (PBS), after which the slides were once again dried at room temperature overnight. To permeabilize the sperm cells, the slides were incubated in 0.1% Triton X-100 and 0.1% sodium citrate in PBS for 2 minutes in 4 °C, washed 3 times in PBS, and allowed to dry for 20 minutes at room temperature. As per the manufacturer protocol, 50 mL of the TUNEL reaction mixture was added onto each slide and incubated for 60 minutes at 37 °C in a humidified atmosphere in the dark and subsequently rinsed in PBS 3 times to remove unbound antibody. 4′,6-Diamidino-2-phenylindole staining solution (MilliporeSigma, Billerica, MA) was added to each slide and incubated at room temperature for 1–5 minutes to fluorescently visualize nuclear DNA. The slides were examined under a fluorescent microscope for positive green fluorescein signals indicating DNA fragmentation. A minimum of 500 spermatozoa were assessed per sample. A SCF of ≥ 15% was considered abnormal on the basis of on previous studies (6).

Ovarian Superovulation, Oocyte Collection, and ICSI

Stimulation protocols were similar between female patients with male partners in the normal, moderate, and severe SCF groups. Patients were treated daily with gonadotropins (Ovidrel [Merck KGaA, Darmstadt, Germany], Gonal-F [EMD-Serono, Geneva, Switzerland], and/or Menopur [Ferring Pharmaceuticals Inc., Parsippany, NJ]). A gonadotropin-releasing hormone antagonist (Ganirelix acetate [Merck, Kenilworth, NJ] or Cetrotide [EMD-Serono Inc., Rockland, MA]) was used to inhibit premature luteinizing hormone surges in controlled ovarian stimulation to prevent precocious ovulation. When at least 2 follicles reached a diameter of ≥ 17 mm, human chorionic gonadotropin (hCG) (Pregnyl Merck) was administered to trigger ovulation. Oocyte retrieval was performed under sedation by a transvaginal sonography-guided technique 35–37 hours after hCG administration. Two hours after oocyte collection, cumulus cells were denudated by 40 IU/mL of recombinant human hyaluronidase (ICSI Cumulase; Cooper Surgical, Inc., Trumbull, CT) (11). Cleaned oocytes were then incubated 1–2 hours before ICSI, which has been previously described (12).

Embryo Evaluation

Grading criteria for human blastocysts have been previously described (13). Briefly, good-quality blastocysts had the following morphological grades according to a grading criterion used in previous studies (14): degree of expansion and hatching status, 1–3 (blastocoele expansion ≥ 50% the volume of the embryo); inner cell mass, A–B (distinct inner cell mass with healthy, compacted cells); and trophectoderm, A–B (healthy cells forming a cohesive epithelium).

Embryo Transfer and Pregnancy Assessment

Starting 1 day after retrieval, patients received intramuscular progesterone supplementation (50 mg daily). Patients with an endometrial lining of at least 7 mm underwent a fresh embryo transfer on day 3 or 5, depending on the developmental characteristics of the embryo. The serum β-hCG levels were measured 10–14 days after retrieval to confirm pregnancy. A clinical pregnancy was defined as ultrasound confirmation of a gestational sac with fetal heart activity 6–8 weeks after transfer.

Statistical Analysis

Descriptive statistics were performed for all clinical outcome variables. A student’s t test or analysis of variance was used to compare continuous variables, and a χ² analysis or the Fisher’s exact test was performed to compare categorical variables. Statistically significant results were defined as P values of ≤ .05.

RESULTS

In this study, a total of 76 couples with elevated SCF were identified with median maternal and paternal ages of 35 (IQR, 32–36 years ) and 40 (IQR, 36–50 years) years, respectively. They underwent a total of 168 ICSI cycles reporting a low fertilization rate of 67.1% and a clinical pregnancy rate of only 16.6%, resulting from an embryo implantation of only 8.5%, in comparison with a general control and a control consisting only of male partners with normal SCF (<15%) (P<.00001). This was further exacerbated by a pregnancy loss that reached 52.9% (P<.00001). Considering that these couples had a mean sperm concentration of 33 × 10⁶/mL, 33% motility, and maternal age comparable with that of the controls, we suspected the presence of a subtle male factor. Therefore, we performed a sperm DNA integrity assessment by TUNEL that evidenced a mean SCF of 21.6%. The semen parameter characteristics separated by the degree of SCF are depicted in Table 1.

TABLE 1.

Semen parameters.

	Control	Control with normal SCF	Moderate	Severe	P value
Men	21,256	737	60	16	—
Volume (mL)	2.8 ± 2.1	2.3 ± 1.4	2.3 ± 2.7	2.3 ± 1.4	NS
Concentration (106/mL)	44.7 ± 35.1	53.0 ± 44.9	33.7 ± 28.4	32.0 ± 33.4	NS
Motility (%)	42.4 ± 17.5	41.2 ± 15.1	36.2 ± 14.8	23.3 ± 17.8	<.01a
Morphology (%)	2.5 ± 1.8	2.5 ± 1.2	2.3 ± 0.9	2.0 ± 0.9	NS
SCF (%)	—	9.7 ± 4.5	20.8 ± 4.3	46.2 ± 20.5	<.0001b

Note: There was a significant difference in motility among the normal, moderate, and severe SCF groups. NS = not significant; SCF sperm chromatin fragmentation.

^aAnalysis of variance, effect of degree of SCF (normal, moderate, severe) on motility.

^bAnalysis of variance, degree of SCF.

In Table 2, the clinical outcome of history cycles before SCF assessment of these couples is depicted according to the level of SCF and compared with the controls. Despite a comparable maternal age, there were impaired fertilization (P<.00001), embryo implantation (P<.00001), and clinical pregnancy (P<.0001) rates in the moderate and severe groups, compared to the normal SCF control. Moreover, the pregnancy loss was evidently increased in both the moderate and severe fragmentation groups compared with that in the control groups (P<.00001).

TABLE 2.

ICSI clinical outcome before sperm treatment.

		SCF
	Control	Control with normal SCF	Moderate	Severe	P value
Couples	21,256	737	60	16	—
Maternal age (M ± SD)	37.2 ± 7.0	34.0 ± 4.3	34.0 ± 3.0	33.2 ± 4.8	NS
Paternal age (M ± SD)	43.8 ± 6.3	39.4 ± 6.0	41.6 ± 8.0	45.2 ± 9.7	NS
Cycles	38,892	1,638	133	35	—
Number of oocytes	408,826	19,076	1,576	382	—
MII oocytes	328,745	14,832	1,281	318	—
Fertilization (%)	247,299/328,745 (75.2)	10,874/14,832 (73.3)	858/1,281 (67.0)	216/318 (67.9)	<.00001a
Embryos transferred (M)	2.0	1.1	1.6	2.0	—
Embryo implantation (%)	17,209/79,088 (21.8)	250/1,780 (14.0)	20/213 (9.4)	4/70 (5.7)	<.00001a
+ β-hCG (%)	14,990/29,861 (50.2)	341/839 (40.6)	30/88 (34.1)	8/26 (30.8)	<.00001a
+ FHB (%)	11,136/29,861 (37.3)	209/839 (24.9)	17/88 (19.3)	2/26 (7.7)	<.00001a
Delivery (%)	9,760/29,861 (32.6)	168/839 (20.0)	8/88 (9.1)	1/26 (3.8)	<.00001a
Pregnancy loss (%)	1,376/11,136 (12.4)	41/209 (19.6)	9/17 (52.9)	1/2 (50)	<.00001a

Note: There was a significant difference in the fertilization, embryo implantation, positive b-hCG, clinical pregnancy, delivery, and pregnancy loss rates among the normal, moderate, and severe SCF groups in their history cycles. β-hCG = human chorionic gonadotropin; FHB = fetal heart beat; NS = not significant; M = mean; MII = metaphase II; SCF = sperm chromatin fragmentation; SD = standard deviation.

^aFisher’s exact test, 3 × 2, 2 df, effect of degree of SCF (normal, moderate, severe) on ICSI clinical outcome in the history cycles.

To address the high SCF status, in a subsequent cycle, couples in the moderate and severe SCF groups were counseled accordingly and offered to undergo surgical sperm retrieval or MFSS. A total of 13 male partners agreed to undergo surgical sperm retrieval. Of these 13 male partners, 5 fell into the moderate SCF category and 8 fell into the severe SCF category. The remaining 63 patients underwent an ICSI cycle using spermatozoa processed by MFSS. All 76 patients underwent an initial ICSI cycle with those interventions. For all couples with elevated SCF (moderate and severe), although no meaningful improvement in fertilization between the history and first post-treatment cycle (493/745, 66.2%) was noted, there was an improvement in the clinical pregnancy (20/51, 39.2%) (P<.01), delivery (19/51, 37.3%) (P<.001), and implantation (27/115, 23.5%) (P<.001) rates in the first post-treatment cycle. The pregnancy loss considerably decreased from 53% (10/19) to 5% (1/20) (P<.01) in the first post-treatment cycle. To further investigate the effect of severity of SCF (moderate vs. severe), when comparing clinical outcomes of the first post-treatment cycle between the moderate and severe SCF groups, the fertilization was significantly lower in the severe group at 64.5% (93/144) (P<.001), without any discernable difference on the remaining clinical outcome (Table 3).

TABLE 3.

ICSI clinical outcome for the first cycle immediately after sperm treatment.

	SCF
	Moderate	Severe	P value
Couples	60	16	—
Maternal age (M ± SD)	35.4 ± 2.7	33.7 ± 5.0	NS
Paternal age (M ± SD)	41.9 ± 7.7	42.6 ± 8.7	NS
Cycles	60	16	—
Number of oocytes	775	194	–
MII oocytes	601	144	–
Embryos transferred (M)	1.4	1.9	–
Fertilization (%)	400/601 (66.6)	93/144 (64.5)	<.001a
Embryo Implantation (%)	20/85 (23.5)	7/30 (23.3)	NS
+ β-hCG (%)	18/38 (47.4)	6/13 (46.2)	NS
+ FHB (%)	15/38 (39.5)	5/13 (38.5)	NS
Delivery (%)	14/38 (36.8)	5/13 (38.5)	NS
Pregnancy Loss (%)	1/15 (6.7)	0/5 (0)	NS

Note: There was a significant difference in the fertilization rate between the moderate and severe SCF groups in the first cycle after sperm treatment in the cycles using microfluidic processed spermatozoa or spermatozoa retrieved from the testis. β-hCG = human chorionic gonadotropin; FHB = fetal heart beat; NS = not significant; M = mean; MII = metaphase II; SCF = sperm chromatin fragmentation; SD standard deviation.

^aFisher’s exact test, 2 × 2, 1 df, effect of degree of SCF (normal, moderate, severe) on ICSI clinical outcome in the first post-treatment cycle.

To identify the effect of our intervention, in Table 4, we depicted the clinical outcomes of all post-treatment ICSI cycles and compared them to those of the history cycles, stratified according to the level of SCF. In the moderate category (SCF, 15%–30%; n × 60), there was a meaningful improvement in the implantation (24.3% vs. 9.4%) (P<.001), clinical pregnancy (40.4% vs. 19.3%) (P<.01), and delivery (36.2% vs. 9.1%) (P<.001) rates, which yielded a reduced pregnancy loss rate of 10.5% from 52.9% (P<.01), compared with that of the history cycles. In the severe SCF category (SCF, >30%; n = 16), there was also a meaningful improvement in the implantation (15.4% vs. 5.7%), clinical pregnancy (21.7% vs. 7.7%), and delivery (21.7% vs. 3.8%) rates, as well as a reduction in the pregnancy loss rate (0 vs. 50.0%), albeit not significant. Ovarian stimulation protocols were similar between couples with male partners in the normal, moderate, and severe SCF groups with mean estrogen levels of 1,923.2 ± 1,016.5 pg/mL in the history cycles and 2,060 ± 1,100.2 pg/mL in the treatment cycles.

TABLE 4.

ICSI clinical outcome before and after sperm treatment, separated by degree of SCF

	Moderate			Severe
	History cycle	Treatment cycle	P value	History cycle	Treatment cycle	P value
Couples	60			16
Maternal age (M ± SD)	34.0 ± 3.0	35.7 ± 2.6	<.0001	33.2 ± 4.0	34.8 ± 4.4	<.001
Paternal age (M ± SD)	41.6 ± 8.0	42.1 ± 7.3	NS	45.2 ± 9.7	45.4 ± 7.1	NS
Cycles	133	93	—	35	31	—
Number of oocytes	1,576	1,171	—	382	332	—
MII oocytes	1,281	921	—	318	254	—
Fertilization (%)	858/1,281 (67.0)	605/921 (65.7)	NS	216/318 (67.9)	172/254 (67.7)	NS
Embryos transferred (M)	1.6	1.2	—	2.0	1.7	—
Embryo implantation (%)	20/213 (9.4)	27/111 (24.3)	<.001a	4/70 (5.7)	8/52 (15.4)	NS
+ β-hCG (%)	30/88 (34.1)	22/47 (46.8)	NS	8/26 (30.8)	9/23 (39.1)	NS
+ FHB (%)	17/88 (19.3)	19/47 (40.4)	<.01a	2/26 (7.7)	5/23 (21.7)	NS
Delivery (%)	8/88 (9.1)	17/47 (36.2)	<.001a	1/26 (3.8)	5/23 (21.7)	NS
Pregnancy loss (%)	9/17 (52.9)	2/19 (10.5)	<.01a	1/2 (50.0)	0/5 (0)	NS

Note: Within the moderate SCF group, the embryo implantation, clinical pregnancy, and delivery rates were significantly higher in treatment cycles than in the history cycles, whereas the pregnancy loss rate was reduced. Albeit not mathematically significant, the same trend was observed in the severe SCF group. β-hCG = human chorionic gonadotropin; FHB = fetal heart beat; NS = not significant; M = mean; MII = metaphase II; SD = standard deviation.

^aFisher’s exact test, 2 × 2, 1 df, effect of sperm treatment on ICSI clinical outcome.

To control for the presence of an eventual confounding female factor, we included 408 couples (median maternal age, 44 [IQR, 45–40 years] years; median paternal age, 44 [IQR, 48–41 years] years). In this cohort, 341 couples had elevated mean DNA fragmentation of 27.6% and used their own oocytes in 757 ICSI cycles, resulting in fertilization, implantation, clinical pregnancy, and delivery rates of 67.1%, 14.4%, 28.0%, and 23.7%, respectively. Of the 408 couples, 67 (SCF of 24.3%) underwent ICSI using donor (21–34 years) oocytes resulting in significantly higher rates of fertilization (78.1%) (P<.00001) and embryo implantation (29.1%) (P<.0001). Although the fertilization and implantation rates were remarkably higher in cases using donor oocytes, the clinical pregnancy rate was only 36.1%, together with a 19.2% pregnancy loss (Table 5). This underpins the deleterious effect of the male gamete even in the utilization of younger and healthier oocytes with a putative lower aneuploidy and superior oocyte repair mechanism.

TABLE 5.

ICSI clinical outcome utilizing donor oocytes or patient oocytes injected with elevated SCF.

	Autologous	Heterologous	P value
Couples	341	67	—
Maternal (recipient) age (M ± SD)	38.4 ± 5.1	42.6 ± 4.3	<.0001
Paternal age (M ± SD)	42.2 ± 8.7	44.2 ± 6.6	<.05
Cycles	757	98	—
Number of oocytes	9,055	1,177	—
MII oocytes	7,187	1,007	—
Mean number of embryos transferred	2.3	1.6
Fertilization (%)	4,824/7,187 (67.1)	786/1,007 (78.1)	<.00001a
Embryo implantation (%)	150/1,042 (14.4)	34/117 (29.1)	<.0001a
+ β-hCG (%)	188/447 (42.1)	41/72 (56.9)	<.05a
+ FHB (%)	125/447 (28.0)	26/72 (36.1)	NS
Delivery (%)	106/447 (23.7)	21/72 (29.2)	NS
Pregnancy loss (%)	19/125 (15.2)	5/26 (19.2)	NS
SCF (%)	27.6 ± 9.6	24.3 ± 8.8	NS

Note: Fertilization, embryo implantation, and positive b-hCG were compromised in the cycles using patients’ own oocytes injected with elevated SCF compared with those in cycles with donor oocytes. No significance was found in the clinical pregnancy, delivery, and pregnancy loss rates. β-hCG = human chorionic gonadotropin; FHB = fetal heart beat; NS = not significant; M = mean; MII = metaphase II; SCF = sperm chromatin fragmentation; SD = standard deviation.

^aχ², 2 × 2, 1 df, effect of elevated SCF on autologous vs. heterologous oocyte injection.

To further substantiate the exclusive contribution of the male gamete genome to embryo development, we controlled for the female gamete contribution by comparing the ICSI clinical outcomes of 32 cycles where male partners with normal SCF (n = 17, 8.9% SCF) used split sibling donor oocytes with those of couples whose male partner had abnormal, uncorrected SCF (n = 12, 23.1% SCF) (Table 6). There was a significant reduction in the fertilization (69.3%) (P<.001), embryo implantation (15.0%) (P<.05), and delivery (15.4%) (P<.05) rates in cycles using spermatozoa with abnormal SCF (uncorrected). Furthermore, we compared the ICSI clinical outcomes between cycles using split sibling donor oocytes of 8 patients with normal SCF (7.1%) and 7 patients with abnormal SCF (24.1%), although, this time, the level of SCF was alleviated to 5.5% because of the use of microfluidic processed spermatozoa (n = 4), testicular spermatozoa (n = 2), or donor spermatozoa (n = 1) (Table 7). This correction affected the clinical outcomes, where a meaningful increase can actually be observed in the clinical pregnancy (42.9 vs. 66.7%), embryo implantation (30.0% vs. 45.5%), and delivery (42.9 vs. 50.0%) rates when using spermatozoa processed by microfluidics or spermatozoa from a different source, albeit not statistically significant. Furthermore, the fertilization rate decreased from 89.4%–73.8% (P<.05) in the cycles using sibling oocytes injected with abnormal, corrected SCF perhaps because of the use of different processing methods.

TABLE 6.

ICSI clinical outcome utilizing sibling donor oocytes injected with spermatozoa with normal or abnormal SCF.

	SCF
	Normal	Abnormal	P value
Couples	17	12	—
Maternal age (M ± SD)	44.2 ± 3.6	42.3 ± 4.4	NS
Paternal age (M ± SD)	44.4 ± 6.8	43.6 ± 6.6	NS
Cycles	17	15	—
Number of oocytes retrieved/frozen	187	137
MII injecteda	157	127	—
Fertilization (%)	137/157 (87.3)	88/127 (69.3)	<.001b
+ β-hCG (%)	8/14 (57.1)	7/13 (53.8)	NS
+ FHB (%)	8/14 (57.1)	3/13 (23.1)	NS
Embryo implantation (%)	9/19 (47.4)	3/20 (15.0)	<.05b
Delivery (%)	8/14 (57.1)	2/13 (15.4)	<.05b
Pregnancy loss (%)	0/8 (0)	1/3 (33.3)	NS
SCF (%)	8.9 ± 3.0	23.1 ± 6.5	<.00001c

Note: There was a significant reduction in the fertilization, embryo implantation, and delivery rates when sibling donor oocytes were injected with abnormal spermatozoa compared with the other half of the split donor oocytes injected with spermatozoa with normal SCF. βhCG = human chorionic gonadotropin; FHB = fetal heart beat; NS = notsignificant; M = mean; MII = metaphase II; SCF = sperm chromatin fragmentation; SD = standard deviation.

^aIncludes only those that survived thawing.

^bFisher’s exact test, 2 × 2, 1 df, effect of SCF on sibling donor oocyte injection.

^cUnpaired Student’s t test, degree of SCF.

TABLE 7.

ICSI clinical outcome using sibling donor oocytes injected with spermatozoa with normal or abnormal SCF corrected by MFSS or testicular sperm extraction.

	SCF
	Normal	Abnormal (corrected)	P value
Couples	8	7	—
Maternal age (M ± SD)	43.1 ± 2.8	44.3 ± 3.9
Paternal age (M ± SD)	42.9 ± 2.6	47.3 ± 7.9
Cycles	8	7	—
Number of oocytes retrieved/frozen	76	85
MII injecteda	66	80	—
Fertilization (%)	59/66 (89.4)	59/80 (73.8)	<.05b
+ βhCG (%)	3/7 (42.9)	5/6 (83.3)	NS
+ FHB (%)	3/7 (42.9)	4/6 (66.7)	NS
Embryo implantation (%)	3/10 (30.0)	5/11 (45.5)	NS
Delivery (%)	3/7 (42.9)	3/6 (50.0)	NS
Pregnancy loss (%)	0/3 (0)	1/4 (25.0)	NS
SCF (%)	7.1 ± 1.8	5.5 ± 1.3c	NS

Note: When sibling donor oocytes were injected with spermatozoa with abnormal SCF but this time corrected by MFSS or testicular surgical retrieval, the ICSI clinical outcome became normalized compared with that of those injected with spermatozoa with normal SCF. β-hCG = human chorionic gonadotropin; FHB = fetal heart beat; NS = not significant; M = mean; MII = metaphase II; SCF sperm chromatin fragmentation; SD = standard deviation.

^aIncludes only those that survived thawing.

^bFisher’s exact test, 2 × 2, 1 df, effect of SCF on sibling donor oocyte injection.

^cPost-treatment SCF, SCF before treatment was 24.1% ± 12.7%.

DISCUSSION

It is extremely puzzling to treat couples where a male partner has adequate semen parameters and a female partner is of a relatively young age with a normal infertility workup. Indeed, semen specimens with parameters somewhat below the threshold indicated by the WHO 2021 still grant successful pregnancies with ICSI (15). Therefore, when such couples have an unexpected poor outcome, the only option left is to repeat a cycle attempting to tweak the stimulation protocol, hoping for a better sequel (16). Our center functions as a tertiary health care facility where couples are referred to carrying the belief and at times even the expectation of achieving the desired results, not to mention the awareness that several health insurance providers cover a limited number of ART cycles (17).

It has become our practice to identify an eventual subtle male factor that may not appear obvious after a semen analysis evaluation (4). Therefore, in our study after treating 76 couples with a surprisingly poor outcome, we decided to assess the SCF by TUNEL. The utility of identifying a subtle male factor has been previously demonstrated to guide further reproductive options to address the compromised SCF when couples with unexplained infertility perform poorly with IUI (4). This can be performed on the acquired knowledge that the SCF is directly correlated with motility (7). Indeed, a sperm preparation that selects for the most motile sperm cells, such as through a microfluidic chamber, will enrich the proportion of male gametes with an intact DNA (18). Another approach, although more invasive and less appealing to a male patient, is the retrieval of spermatozoa directly from the epididymis or testis. Indeed, it has been observed that the SCF increases as the spermatozoa travel through the male genital tract to reach an ejaculate (5). Hence, when the surgically retrieved spermatozoa from these men are used to inseminate the oocytes of their female partner, superior clinical outcome was observed (5).

Therefore, when we recorded the initial cycle after sperm treatment, whether by microsurgical sperm retrieval or processed by microfluidics, we noticed an outstanding improvement in SCF and consequent clinical outcome results, with an improvement in the clinical pregnancy rate, irrespective of the degree of SCF, from 16.6% in the history cycle to 39.2% in the post-treatment cycle (P<.01). This amelioration persisted throughout all subsequent cycles performed after SCF assessment; the clinical pregnancy increased to 34.2% in the post-treatment cycles from 16.6% in the history cycle (P<.01). This indicated that in a selected patient population, the male factor cannot be fully identified by measuring volume, concentration, or motility, and unfortunately even by careful assessment of the morphology of the male gamete (19).

More conservative approaches have been proposed entailing the introduction of healthy lifestyle modifications (20) or the adoption of antioxidants (21) with the claim of lowering the proportion of SCF in the semen sample. However, results with these treatments have been inconsistent and, at best, inconclusive (22).

To evidence the exclusive role of a suboptimal male gamete genome on embryo development and to correct for an eventual confounding female factor, we have identified couples with a male partner displaying an elevated SCF and a female partner who used their own and donated healthy oocytes. We found only modest improvement in the clinical pregnancy rate from 28.0% where patient oocytes were injected with male partner spermatozoa with elevated SCF, to 36.1% where donor oocytes were injected with male partner spermatozoa with a similar elevated SCF. This was investigated in an earlier study by Esbert et al. (23), which observed that the in vitro fertilization or ICSI clinical outcome was unaffected when spermatozoa with elevated DNA fragmentation were used to inject their own female partner or donor oocytes. This observation was confirmed in a later study from the same group, where patients’ own oocytes and donor oocytes were injected with high or low SCF, and in the donor cohort, the injection of spermatozoa with high SCF indicated a clear delay in embryo kinetics (24).

To undoubtfully remove the female gamete contribution, we examined the effect of injecting sibling donor oocytes split between different couples who used spermatozoa of their respective male partner with normal or elevated SCF. Unfortunately, there was a clear reduction in embryo implantation (P<.05) and pregnancy outcome (P<.05) when SCF was elevated. This impairment was revoked when male partner spermatozoa with high SCF were processed by MFSS or retrieved directly from the testicle. Our findings are corroborated by a study that executed a similar design where sibling oocytes split between different couples were inseminated by their respective male partner characterized as low SCF (8.5% ± 3.7%) or high (27.6% ± 8.3%) and resulted in a significantly lower implantation rate in the latter (25). This confirms that a fragmented male genome is capable of negatively impacting embryo development in spite the use of healthy, young donor oocytes, and overwhelming the putative ooplasmic repair mechanism.

In this study, we were able to reveal a subtle male factor in patients who otherwise had a semen sample adequate for ICSI. The cohort included in the study is relatively small and entails a selected patient population on the basis of unexpectedly poor clinical outcome. Moreover, the sperm DNA fragmentation assay provides only the rate of DNA breakage and, therefore, cannot guarantee that a single specific spermatozoon with an unfragmented genome be selected to fertilize an oocyte.

CONCLUSION

In conclusion, the sophistication and versatility acquired in ARTs, particularly with the introduction of ICSI, grants the ability to treat most aspects of human infertility. However, it remains complicated to manage couples with idiopathic infertility, particularly when a female partner is of a relatively young age and a male partner’s semen parameters do not justify the basis for the reproductive failure. In these cases, the identification of a hidden and more intrinsic male factor should be attempted, and this can be initiated by assessing the chromatin integrity of the male gamete. Once a high SCF has been confirmed, certain amendments can be introduced aimed to select spermatozoa with high DNA integrity, such as those retrieved proximally from the male genital tract or processed through a microfluidic chamber. This correction allows superior clinical outcome, allowing couples to fulfill their childwish.