Edmund Wojcicki (plaintiff), acting individually and as the executor of the estate of his wife, Sherry Wojcicki (decedent), brought this action for medical malpractice and loss of consortium, alleging negligence by Dr. Joan Caragher (defendant) in treating the decedent for stroke. After a trial in the Superior Court, the jury returned a verdict in favor of the defendant. The plaintiff filed a motion for a new trial, claiming that an expert witness for the defense had provided false and misleading testimony, and asked the judge to impose sanctions. After extensive posttrial proceedings, including a deposition of the expert witness, the motion judge, who also had presided at trial, granted the plaintiff’s motion and imposed sanctions against both the expert and the defendant. A single justice of the Appeals Court granted the defendant’s interlocutory appeal from these orders, and stayed the proceedings in the Superior Court pending appellate review. We granted the plaintiff’s application for direct appellate review, and we now reverse.
1. Background. The decedent was diagnosed with cancer in April, 1999, and received the final dose of chemotherapy to shrink a large, inoperable tumor in her right breast on July 1, 1999. At home the next day, at approximately 11:30 a.m., the plaintiff discovered the decedent collapsed on a couch, unable to move her left arm or leg. The decedent quickly was taken to a hospital by ambulance; when she arrived in the emergency department, she appeared alert, complaining of headache, left-sided weakness, and slurred speech. After examining the decedent and performing the relevant diagnostic tests, including blood work and a CT scan, hospital staff determined that she had suffered an ischemic stroke.2
Treatment options for ischemic stroke were very limited until 1995, when a study conducted by the National Institute of *202Neurological Disorders and Stroke (NINDS study) demonstrated that tissue plasminogen activator (t-PA), a drug that can dissolve blood clots, may reduce the long-term disability that often follows ischemic stroke. Because of the serious risks associated with t-PA use, including fatal intracranial bleeding, the drug had to be administered within three hours of the onset of stroke, and the patient had to undergo a CT scan before the t-PA could be administered, to rule out any bleeding in the brain. The researchers also developed exclusion criteria to ensure that patients with a heightened risk of intracranial bleeding did not receive the drug.3 The Food and Drug Administration (FDA) approved the use of t-PA to treat ischemic stroke in 1996.
When the defendant, the attending physician in the emergency department, learned of the decedent’s recent cancer diagnosis and treatment, she telephoned the decedent’s oncologist to determine whether her cancer was metastatic; if the cancer had spread to the decedent’s brain, she would face an increased risk of intracranial bleeding and therefore would not be an appropriate candidate for t-PA. The oncologist informed the defendant that he did not believe the decedent’s breast cancer to be metastatic, but the testing necessary to determine whether the cancer had reached her brain had not been completed.4 The defendant ultimately decided that the decedent was not an appropriate candidate for t-PA.
Following the defendant’s decision not to administer t-PA, *203the decedent was transferred to another hospital.5 Two weeks later, she was moved to a rehabilitation facility, where she remained for several weeks. The stroke left the decedent paralyzed on the left side of her body. Although surgery to remove the tumor in her breast was successful, she died from cancer-related complications in 2002.6
The plaintiff filed this action for medical malpractice in 2000, claiming that the decedent was an appropriate candidate for t-PA therapy on July 2, 1999; that the defendant deviated from the standard of care by failing to use this drug to treat her stroke, or to transfer the decedent in time to receive t-PA at another hospital; and that the failure to administer t-PA caused the decedent’s medical outcome. At trial, the plaintiff introduced an article from the New England Journal of Medicine summarizing the results of the 1995 NINDS study; two expert witnesses who testified on behalf of the plaintiff stated that this article established the standard of care regarding the use of t-PA to treat stroke patients in 1999. These experts testified that a patient like the decedent would be a suitable candidate for t-PA therapy because she satisfied all of the inclusion criteria set out in the article — t-PA could have been administered within three hours of the onset of stroke, and the CT scan did not reveal any bleeding in her brain — and because she did not present with any exclusion criteria. Both experts also stated that cancer is not a contraindication to the use of t-PA unless it has metastasized to the patient’s brain.
The defendant, testifying at trial, explained that she did not consider the decedent to be an appropriate candidate for t-PA therapy because she recently had been diagnosed with breast cancer; she had undergone chemotherapy the day before the stroke; and the testing necessary to determine whether her cancer had metastasized had not been completed. Each of these *204factors, the defendant claimed, increased the decedent’s risk of intracranial bleeding. The defendant offered the testimony of three expert witnesses to support her position. One of these experts, Dr. Fred Hochberg, a neurologist specializing in the treatment of cancer patients, testified that no cancer patients were included in the NINDS study.7 On cross-examination, Hochberg insisted that, although the article from the New England Journal of Medicine did not mention cancer, “the actual study” contained “zero” cancer patients.8 When defense counsel later asked Hochberg to clarify the difference between the study and the published article, Hochberg explained that all data pertaining to patients enrolled in the study were contained on *205two CD-ROMs, available from the National Institutes of Health, and that some of this data were not included in the article.9 Hochberg also acknowledged that cancer was not listed as an exclusion criteria in the NINDS study.
The jury found that the defendant had not been negligent in treating the decedent. After the verdict, the judge, accompanied by counsel, met briefly with jurors; one juror mentioned that she was particularly impressed by Hochberg’s testimony.10 Later that day, the plaintiff’s counsel contacted one of his own experts to discuss his concerns about Hochberg’s testimony. That expert informed him that Hochberg’s claim that zero cancer patients participated in the NINDS study was not supported by the data. Counsel then contacted the NINDS study coordinator and discovered that fifty-nine of the 624 participants reported having been diagnosed with a malignancy at some point in their lives.
Based on this information, the plaintiff filed a motion for a *206new trial pursuant to Mass. R. Civ. R 59 (a), 365 Mass. 827 (1974), and a motion for relief from judgment pursuant to Mass. R. Civ. R 60 (b) (3), 365 Mass. 828 (1974), arguing that Hochberg gave false and misleading testimony about a central issue in the case when he testified that no cancer patients were included in the NINDS study. The plaintiff characterized this testimony as perjury, claimed that its admission constituted fraud on the court, and asked the judge to impose sanctions. The plaintiff submitted affidavits in support of the motion from the study’s lead biostatistician, averring that fifty-nine study participants reported a previous cancer diagnosis, and that this information was available on the CD-ROM from the National Technical Information Service.
At the conclusion of the first hearing on the motion, the judge ordered Hochberg to appear for a deposition regarding the basis of his testimony about the NINDS study. Hochberg testified repeatedly at the deposition that his trial statements about “cancer patients” referred only to patients like the decedent, with acute, recently diagnosed, active cancer. He further testified that, although he had never reviewed the data on the CDROMs, he based his statement that patients with active cancer were excluded from the NINDS study on a telephone conversation with the lead biostatistician.11
The plaintiff responded with a supplemental affidavit from the biostatistician, in which she explained that neither the published article nor the data on the CD-ROMs contained any information about whether the participants who. reported a diagnosis of malignancy had active cancer at the time of the study. The only way to determine the status of cancer in those fifty-nine patients during the study, she later explained, would be to identify the patients, contact their treating physicians, and review their individual medical records. She also denied ever having a telephone conversation with Hochberg, and produced *207documents proving she was out of State when he claimed their conversation occurred.
At the next hearing on the motion, the judge found that Hochberg intended to create the impression that the CD-ROMs indicated that no cancer patients participated in the NINDS study when, in fact, he did not review the data and lacked such knowledge. Accordingly, she ordered a new trial and awarded the costs of the first trial to the plaintiff as sanctions. Because she imposed sanctions, the judge gave Hochberg the opportunity to be heard before issuing her final, written decision.
Hochberg obtained his own counsel, and the posttrial proceedings continued over several months. Hochberg presented evidence that researchers considered “a serious medical illness that is likely to interfere with” the NINDS study to be an exclusion factor. He deposed the biostatistician and obtained affidavits from three of the study’s principal investigators, attesting that, to the best of their knowledge, none of the participants enrolled through their hospitals had active cancer.12 Finally, counsel for the defendant submitted an affidavit in which he confirmed that he told Hochberg before trial that he did not need to purchase the CD-ROM. Defense counsel also explained that, at trial, he was careful to ask Hochberg only questions about patients who “presented with” cancer, a clinical term meaning active cancer, and that he intentionally did not ask Hochberg if he had reviewed the data on the CD-ROMs, because he knew that Hochberg had not done so.
*208On November 16, 2004, the judge issued her written decision on the plaintiff’s motion for a new trial, finding by clear and convincing evidence that Hochberg intentionally testified at trial about the existence of a fact (the exclusion of cancer patients from the NINDS study) that he did not know to be true; “deliberately created the false impression . . . that he had reviewed the CD-ROM . . . and that it showed that no cancer patients were included in the study”; and testified falsely in his posttrial deposition that he had a telephone conversation with the lead biostatistician. The judge found that the evidence presented in the plaintiff’s motion was newly discovered because the defendant did not disclose that Hochberg would testify about the characteristics of study participants, specifically, that no cancer patients were included in the study, and because the study data on CD-ROM could be read only by a statistician using special software. These circumstances, she concluded, warranted a new trial. In addition, while acknowledging that she could find no clear precedent authorizing an award of sanctions against a nonparty witness, the judge imposed sanctions against Hochberg for this false testimony. The judge also ordered sanctions against the defendant, based on defense counsel’s failure to disclose that Hochberg would testify that no patients with active cancer were included in the study, his failure to correct Hochberg’s testimony that “zero” cancer patients were included in the study, and his role in eliciting testimony that created the false impression that Hochberg had reviewed the CD-ROMs.13
2. Discussion. The sole issue raised on appeal is the propriety of the order allowing the plaintiff’s motion for a new trial and imposing sanctions on Hochberg and the defendant. Rule 59 (a) (1) permits a judge to order a new trial “in an action in which there has been a trial by jury, for any of the reasons for which new trials have heretofore been granted in actions at law in the courts of the Commonwealth.” Rule 60 (b) (3) specifi*209cally authorizes relief from a final judgment for “fraud (whether heretofore denominated intrinsic or extrinsic), misrepresentation, or other misconduct of an adverse party.” The resolution of motions for a new trial and motions for relief from judgment rests in the discretion of the trial judge, and we review orders granting such relief for an abuse of that discretion.14 E.g., Gath v. M/A-Com, Inc., 440 Mass. 482, 492 (2003) (motion for new trial); Cullen Enters., Inc. v. Massachusetts Prop. Ins. Underwriting Ass’n, 399 Mass. 886, 894 (1987) (motion for relief from judgment)-. We similarly review orders imposing sanctions for an abuse of discretion. Ayash v. Dana-Farber Cancer Inst., 443 Mass. 367, 401, cert. denied sub nom. Globe Newspaper Co. v. Ayash, 126 S. Ct. 397 (2005), and cases cited.
The defendant and Hochberg contend that the judge abused her discretion in ordering a new trial and imposing sanctions because they did not commit fraud on the court or provide false or misleading trial testimony. They argue that Hochberg’s testimony about the MINDS study cannot be considered newly discovered evidence, and that a new trial is not the proper remedy in these circumstances. The plaintiff, however, maintains both that Hochberg and defense counsel committed fraud on the court, and that a new trial would be warranted (in the interests of justice) even without such a finding, because Hochberg intentionally gave false and misleading testimony related to a central issue in the case. The plaintiff claims that the evidence that fifty-nine MINDS study participants had some history of cancer is newly discovered, and that the award of sanctions against Hochberg and defense counsel was fair and reasonable.
a. Fraud on the court. “A ‘fraud on the court’ occurs where ‘it can be demonstrated, clearly and convincingly, that a party *210has sentiently set in motion some unconscionable scheme calculated to interfere with the judicial system’s ability impartially to adjudicate a matter by improperly influencing the trier or unfairly hampering the presentation of the opposing party’s claim or defense.’ ” Paternity of Cheryl, 434 Mass. 23, 35 (2001), quoting Rockdale Mgt. Co. v. Shawmut Bank, N.A., 418 Mass. 596, 598 (1994). The doctrine is limited to “that species of fraud which does, or attempts to, defile the court itself, or is a fraud perpetrated by officers of the court so that the judicial machinery can not perform in the usual manner . . . .” Pina v. McGill Dev. Corp., 388 Mass. 159, 165 (1983), quoting Lockwood v. Bowles, 46 F.R.D. 625, 631 (D.D.C. 1969). “Courts have found fraud upon the court only where there has been the most egregious conduct involving a corruption of the judicial process itself. Examples are bribery of judges, employment of counsel to ‘influence’ the court, bribery of the jury, and the involvement of an attorney (an officer of the court) in the perpetration of fraud.” MacDonald v. MacDonald, 407 Mass. 196, 202 (1990), quoting Lockwood v. Bowles, supra at 631-632.
Although it is unclear whether the judge found that Hochberg committed fraud on the court, we conclude that the evidence cannot support such a finding.15 Aspects of Hochberg’s testimony are disturbing, particularly his trial testimony about the CD-ROMs, suggesting that he was familiar with the data, and his posttrial testimony about his conversation with the bio-statistician, which the judge expressly found to be false. However, even “[pjerjury does not constitute ‘fraud upon the court’ ” when there is no evidence that the judicial process itself was corrupted. Lockwood v. Bowles, supra at 628. “The possibility of a witness testifying falsely is always a risk in our judicial process, but there are safeguards within the system to *211guard against such risks. The most basic of these is cross-examination of witnesses . . . .” Id. at 632-633.
The plaintiff also suggests that defense counsel committed fraud on the court because he “launched the fraud” perpetrated by Hochberg when he inquired about the inclusion of cancer patients in the study, knowing that Hochberg had not reviewed the study data. Again, however, the judge did not explicitly find that counsel’s conduct amounted to fraud on the court, and we conclude that the record does not support such a finding. In fact, the judge found that Hochberg told defense counsel he had discussed the exclusion of patients with active cancer with an investigator with the National Institutes of Health, giving counsel a legitimate basis for inquiring about that issue at trial.16
b. False and misleading testimony. While arguing that Hochberg and defense counsel did commit fraud on the court, the plaintiff maintains that the judge ordered a new trial based not on a finding of fraud on the court, but on Hochberg’s false and misleading testimony that “no cancer patients” were included in the study, when in fact, fifty-nine study participants had been diagnosed with cancer at some point during their lives. The defendant and Hochberg, however, counter that Hochberg’s testimony was not false or misleading because he did not testify that no patients with a history of cancer participated in the study, only that no patients “presenting with” cancer, a clinical term meaning active cancer, were included. They emphasize that Hochberg never testified that he reviewed the CD-ROMs, and that he based his testimony regarding the exclusion of *212patients with active cancer from the study on his knowledge, training, and experience.
There is ample support for the judge’s finding that Hochberg’s trial testimony was misleading. Both defense counsel and Hochberg claim that the distinction between active cancer and historical cancer is critical to understanding Hochberg’s testimony, yet neither explained this difference, or even used these terms, at trial. In addition, Hochberg’s description of the CD-ROMs and their contents implied familiarity with the study data, but he had never seen the CD-ROMs or reviewed the data. Therefore, he could not know with certainty that cancer patients were excluded from the study, but he testified without qualification that no cancer patients participated. Indeed, even had he reviewed the CD-ROMs, he would not have been able to determine whether patients with active cancer were included in the study, because the CD-ROMs did not contain information about the cancer status of the participants who reported a diagnosis of malignancy. Like the judge, we are left with significant concerns about the basis for Hochberg’s testimony about the exclusion of cancer patients from the NINDS study.17
While these aspects of Hochberg’s testimony were misleading, it is not clear whether, and to what extent, the judge found that his trial testimony was false. The judge found that Hochberg intentionally created the false impression that he had reviewed the study data, and that he testified to facts that he did not know to be true, but the only “testimony” that she actually found to be “false” was his posttrial testimony about the telephone conversation with the biostatistician.18 Hochberg’s statement that no cancer patients were included in the NINDS study appears to be baseless, but it is difficult, on this record, to conclude that it was false: the evidence adduced posttrial does not establish that any patients with active cancer participated in the study, although fifty-nine patients with some history of cancer did.
*213c. Newly discovered evidence. Even if we assume that Hochberg’s trial testimony was false, however, the plaintiff must demonstrate that the evidence presented in his motion for a new trial was newly discovered. Evidence is considered “newly discovered” in this context only if it was “unknown and unavailable at the time of trial despite the diligence of the moving party.” Leavitt v. Mizner, 404 Mass. 81, 89 (1989), quoting Commonwealth v. Williams, 399 Mass. 60, 64 (1987). DeLuca v. Boston Elevated Ry., 312 Mass. 495, 497 (1942).
The judge found that the plaintiff satisfied his burden of showing that the evidence relied on in support of his motion — the data showing that fifty-nine study participants reported a history of malignancy — was newly discovered because the defendant did not disclose that Hochberg would testify about the characteristics of study participants, and in particular, that he would testify that no cancer patients were included. She also considered relevant the fact that the CD-ROM could be read only using specialized statistical software.
The defendant and Hochberg challenge this conclusion, maintaining that the data cannot be considered newly discovered because the plaintiff should have anticipated that the defendant would dispute the study’s application to cancer patients, and could have obtained the data before or during trial. We agree. The plaintiff introduced the MINDS study and had access to the CD-ROMs containing the underlying study data. Thus, the data were not unavailable, and the plaintiff could have discovered that fifty-nine study participants previously had been diagnosed with cancer. Furthermore, the pretrial disclosures provided that Hochberg “is expected to testify that [the decedent] had many contraindications to thrombolytic therapy, including her anemia, her recent chemotherapy and the possibility of metastatic disease, possibility of a tumor, and the risk of intracerebral bleeding.” This was sufficient to notify the plaintiff that the defendant would argue that the decedent was not an appropriate candidate for t-PA for reasons related to her cancer and its treatment. Because the plaintiff used the MINDS study to establish the standard of care for the administration of t-PA, it was foreseeable that the defendant, through her experts, could challenge the application of that research to cancer patients like *214the decedent. The plaintiff cannot now claim to be unfairly surprised by expert testimony about the relevance of the NINDS study to cancer patients.
More important, even if the plaintiff did not anticipate that a defense expert would testify about the NINDS study data or the exclusion of cancer patients, he could have taken remedial action at trial. The plaintiff did not attempt to explore the basis of Hochberg’s testimony about the study data through cross-examination, the traditional means of attacking the validity of expert testimony. See, e.g., Commonwealth v. Sands, 424 Mass. 184, 186 (1997) (“basis of [expert] testimony can be effectively tested through cross-examination and rebuttal evidence”); Sacco v. Roupenian, 409 Mass. 25, 30 (1990) (“adverse party is free to attack inconsistencies or omissions in the factual foundation, as well as flaws in the expert’s analytical process . . . through cross-examination”). The plaintiff also could have objected and moved to strike Hochberg’s testimony if he was not prepared to counter such testimony based on the pretrial disclosures, or he could have requested a voir dire hearing or a brief continuance for further investigation. Commonwealth v. Martin, 424 Mass. 301, 306 n.1 (1997). Commonwealth v. Pyne, 35 Mass. App. Ct. 36, 39-40 (1993). However, “[t]here was no objection, register of surprise, or request for a continuance by the [plaintiff]. In these circumstances, the [plaintiff]’s claim of lack of notice rings hollow.” Commonwealth v. Cortez, 438 Mass. 123, 126 (2002). See Beaupre v. Cliff Smith & Assocs., 50 Mass. App. Ct. 480, 486-487 (2000) (fact that party did not make specific objection to expert testimony, or request continuance or voir dire of expert, subverts claim of unfair surprise or prejudice).
Indeed, the plaintiff did not take any action regarding Hochberg’s testimony, which he claims was both unanticipated and devastating to his case, until after the jury returned an unfavorable verdict, when the plaintiff’s counsel contacted his own expert to discuss his concerns about Hochberg’s testimony. It took this expert just one hour to confirm his belief that no study data supported Hochberg’s testimony about the exclusion of cancer patients.19 The plaintiff’s appellate counsel advances no reason for trial counsel’s delay in contacting this expert to *215discuss Hochberg’s testimony. We are left with the conclusion that the plaintiff “is attempting to raise now on a motion for a new trial questions which he might have raised but did not present at his trial.” Commonwealth v. Richardson, 361 Mass. 661, 663, cert, denied, 409 U.S. 884 (1972). Given the “public interest in the finality of judgments,” a motion for a new trial should not be granted when the issues raised therein could have been addressed during the trial. Commonwealth v. Tucceri, 412 Mass. 401, 406 (1992). See Davis v. Boston Elevated Ry., 235 Mass. 482, 496 (1920).
Moreover, newly discovered evidence does not warrant a new trial unless that evidence also would “be a real factor with the jury in reaching a decision.” Id. The new evidence must be “material not only in the sense that it is relevant and admissible but also in the sense that it is important evidence of such a nature as to be likely to affect the result.” DeLuca v. Boston Elevated Ry., 312 Mass. 495, 497 (1942). Although the judge need not be convinced that the new evidence would result in a different verdict, it must nevertheless be “important evidence” that “would have genuine effect” on the jury. Davis v. Boston Elevated Ry., supra.
The judge found that Hochberg’s testimony concerned “an issue that was central to the case,” but made no other findings related to its materiality. The plaintiff argues that Hochberg’s testimony affects the very foundation of his case because it made the NINDS study, which established the standard of care for the administration of t-PA, irrelevant to the decedent.20 Citing the juror’s posttrial comment as proof of the impact of Hochberg’s testimony, the plaintiff claims that the fact that some study participants reported having been diagnosed previously with cancer would make a difference at a new trial.
*216However, given the evidence uncovered posttrial, this seems unlikely.21 If there were another trial, the jury would learn that fifty-nine study participants had been diagnosed with cancer at some point in their lives. Presumably, however, the jury also would learn that patients with serious medical illnesses were excluded from the study, and that several physicians involved with the study believed that no patients with active cancer participated at their hospitals.22 Furthermore, the jury would hear expert testimony that there are risks associated with administering t-PA to cancer patients; t-PA should never be used to treat patients with brain metastases; and the testing to determine whether the decedent’s cancer had metastasized had not been completed — all evidence supporting the verdict for the defendant. Finally, the plaintiff extracted a significant concession from Hochberg at trial, namely that cancer was not listed as an exclusion factor in the MINDS study. “[A] new trial ‘ought not to be granted unless on a survey of the whole case it appears to the judge that otherwise a miscarriage of justice would result.’ ” Spiller v. Metropolitan Transit Auth., 348 Mass. 576, 580 (1965), quoting Nicholas v. Lewis Furniture Co., 292 Mass. 500, 507 (1935). The mere possibility that allegedly newly discovered evidence might affect the outcome does not require a new trial. DeLuca v. Boston Elevated Ry., 312 Mass. 495, 500 (1942).
d. Sanctions. After allowing the plaintiff’s motion for a new trial, the judge imposed sanctions against the defendant and Hochberg to compensate the plaintiff for expenses incurred at the first trial and during posttrial proceedings. Sanctions may be ordered “to compensate the aggrieved litigant for the actual loss incurred by the misconduct of the offending party.” Avelino-Wright v. Wright, 51 Mass. App. Ct. 1, 5 (2001), citing Clark v. *217Clark, 47 Mass. App. Ct. 737, 744-745 (1999). For the reasons explained above, however, no new trial is required; therefore, there is no longer a need to compensate the plaintiff for costs associated with a second trial.23
3. Conclusion. For the foregoing reasons, we conclude that the judge abused her discretion in granting the plaintiff’s motion for a new trial and imposing sanctions against the defendant and Hochberg. We accordingly vacate the orders of the judge, reinstate the jury verdict, and enter a final judgment of dismissal.
So ordered.
A stroke is caused by a sudden interruption of blood flow to the brain. An ischemic stroke is caused by the blockage, typically by a blood clot, of a blood vessel in the brain. A hemorrhagic stroke, the less common type, is caused by the sudden rupture of a blood vessel in the brain.
For example, patients were not considered appropriate candidates for t-PA therapy, and therefore were not eligible to participate in the NINDS study, if they had experienced a stroke or serious head trauma within the preceding three months; they had undergone surgery within the previous two weeks; they had a history of intracranial bleeding; or they had high blood pressure or a blood clotting disorder. Stroke patients with symptoms deemed mild or rapidly improving also were not eligible to participate in the study because the risks associated with t-PA use would outweigh its potential benefits for those patients.
The CT scan performed shortly after the decedent’s arrival in the emergency department likely would not have identified brain metastases: the experts who testified at trial agreed that generally an MRI or an enhanced CT scan are the best methods of detecting metastatic cancer. The defendant could not perform these tests in the emergency department, and no neuroradiologist was available at the hospital on July 2, 1999, to review the unenhanced CT scan results.
The decedent arrived at that hospital more than three hours after the onset of her stroke and therefore was not eligible for t-PA therapy there.
It is not clear from the record whether the decedent in fact had metastatic cancer when she suffered the stroke on July 2, 1999. However, eight of fourteen lymph nodes, sampled when the decedent’s tumor was removed later that summer, tested positive for cancer, and the plaintiff testified that the decedent died as a result of metastatic disease, after the cancer spread to her bones.
On direct examination, Hochberg testified to the following:
Q:. “Dr. Hochberg, are you familiar with the NINDS study that was published in December of 1995?”
A.: “Yes.”
Q.: “And as part of this study, Dr, Hochberg, based upon your review of it together with your knowledge, training, and experience, as a neurologist, were there any patients included in that study who presented with cancer?”
A.: “No.”
Q.\ “Were there any patients who are part of that study who presented with breast cancer?”
A.: “There were no patients with breast cancer in the NINDS, either part one or part two of that study.”
The following exchange occurred during Hochberg’s cross examination:
Q.: “How many patients in the NINDS Study had cancer?”
A.: “Zero.”
Q.: “No one knows.”
A.: “Excuse me, zero.”
Q.: “You’re telling the ladies and gentlemen of this jury that this study indicates how many patients had cancer?”
A.: “That’s correct.”
Q.: “Okay. Let’s —”
A.: “It’s not published in there.”
*205Q.: “— well, let’s have a look at it, Doctor. Exhibit 12 is a copy of the NINDS Study. Can you tell us where in that study it states that none of the patients had cancer?”
A.: “You’re referring to the study or the publication of the study?”
Q.: “What is it that you’re holding in your hand, sir?”
A.: “This is the publication of the study. You asked me about the actual NINDS Study, and the answer is zero. The publication makes no mention of the cancer patients. The actual study contained no cancer patients.”
On redirect examination, defense counsel asked Hochberg, “[W]hat is the difference between the publication of the study and the study results themselves?” Hochberg responded: “The NINDS Study was completed in approximately 1994. The current study data, which include the data of every patient who was entered into the study, those who were included, those who were not included ... all are included in two large cd-roms that are available from the National Institutes of Health. To publish the results of the study in the New England Journal of Medicine requires that those data be distilled down to an article that can be read by physicians and by non physicians alike, that will give them the bare information that is logical about the nature of this study. But the actual study details are not embedded in this article.”
According to an affidavit from plaintiff’s counsel, this juror said that she was “most impressed by Dr. Hochberg,” and “after he testified that no cancer patients had been included in the NINDS study, ‘that was the end of the case as far as [she] was concerned.’ ”
Hochberg testified that he telephoned the biostatistician and asked whether a patient like the decedent, who had active breast cancer, the extent of which was not known, and was receiving chemotherapy, would be included in the study. He claimed that she said no, and that he did not ask any other questions. Hochberg’s telephone records revealed a thirty-second telephone call to the biostatistician’s office.
EIeven physicians from eight hospitals throughout the United States served as principal investigators for the NINDS study. Hochberg obtained affidavits from three of the eleven principal investigators. One of the three stated that six study participants from his hospital reported a previous diagnosis of malignancy. At his direction, a data coordinator reviewed the medical records of these patients, and none of them had active cancer during the study. The affidavits provided by the two other principal investigators do not indicate whether anyone reviewed the medical records of the participants enrolled through their hospitals who reported a previous diagnosis of malignancy: their affidavits state only that, to the best of their knowledge and recollection, no patients with active cancer from their hospital participated. Hochberg also obtained an affidavit from a physician on the NINDS data safety monitoring committee, who reported that, to the best of his knowledge and recollection, no patients with acute or active cancer, receiving chemotherapy, were included in the research because they “would have been excluded as a result of an acute medical condition.”
The judge ordered that the defendant pay the plaintiff’s costs and attorney’s fees for services “wasted” during the first trial, and for costs incurred at some of the posttrial proceedings, totaling $68,380. Hochberg was ordered to pay the plaintiff’s costs and attorney’s fees for the remainder of the post-trial proceedings, as well as the costs associated with empanelling a jury for the new trial, totaling $20,305.
The judge did not specify whether she granted relief under Mass. R. Civ. P. 59 (a), 365 Mass. 827 (1974), or under Mass. R. Civ. P. 60 (b), 365 Mass. 828 (1974). The plaintiff argued for relief under both rules in the same document and on essentially the same grounds: that Hochberg, assisted by defense counsel, gave false and misleading testimony on an important issue at trial. It appears that the judge granted relief under rule 59: her memorandum and order refer to the plaintiff’s request for relief as a motion for a new trial, and the plaintiff filed the motion within ten days of the judgment, as required by rule 59.
In the judge’s memorandum and order, she noted that “[a] finding of fraud on the court must be shown by clear and convincing evidence,” and she proceeded to find clear and convincing evidence that Hochberg intentionally testified to a fact without knowing it to be true, deliberately misled the jury, and lied during his posttrial deposition. However, the judge did not discuss or cite any cases dealing with fraud on the court, nor did she explicitly find that a fraud on the court occurred in this case.
The specific findings in the judge’s memorandum and order relating to Hochberg’s misconduct likewise do not support a finding of fraud on the court. She found that defense counsel helped create the false impression that Hochberg had reviewed the study data, and that he failed to correct Hochberg’s testimony that “zero” cancer patients participated in the study. This, however, does not constitute a sentient scheme to defraud the court, particularly where defense counsel had reason to believe that Hochberg had talked to an investigator from the National Institutes of Health about the exclusion of patients with active cancer. The judge also found that defense counsel made inadequate pretrial disclosures about the testimony expected from Hochberg, but “[cjonduct such as nondisclosure to the adverse party or the court of facts pertinent to the matter before it, without more, does not constitute a fraud on the court . . . .” Sahin v. Sahin, 435 Mass. 396, 406 (2001).
On appeal, the defendant and Hochberg argue that Hochberg’s testimony was based on his education, training, and experience. When Hochberg was deposed posttrial, he insisted that his testimony was based on a telephone conversation with the biostatistician. However, he never mentioned this telephone call at trial.
We express no opinion as to whether this testimony constituted perjury.
After discussing the matter with his own expert, counsel contacted the *215NINDS project officer to confirm that cancer patients were not excluded from the study. Counsel contacted the project officer on a Friday afternoon, and was informed on the following Tuesday that fifty-nine participants reported some history of malignancy.
Although the plaintiff now claims that the NINDS study was a critical piece of evidence at trial, we note that neither attorney mentioned the study during opening statements or closing arguments. Although defense counsel referred to Hochberg’s testimony once during his closing, and the plaintiff referred to his testimony several times, these references did not concern his testimony about the study.
The judge alluded to this during a posttrial hearing when she remarked that “one of the most disturbing things about this is the fact that an honest, legitimate defense is likely to have been successful in this case.” She expressed a similar sentiment in her memorandum and order, noting that “a retrial is by no means certain to result in a verdict in favor of plaintiff.”
The plaintiff has not produced any evidence showing that patients with active cancer were included in the study: he relies on the fact that the MINDS data reveal only that fifty-nine participants had been diagnosed with cancer at some point, and suggests the status of their cancer during the research will never be known.
Because we vacate the order imposing sanctions against Hochberg and the defendant, we need not address Hochberg’s contention that the judge lacked the authority to impose sanctions against a nonparty witness.